Current & Future State of Voice 🔊 Technology || C# Corner

Name: Current & Future State of Voice 🔊 Technology || C# Corner | Open Video
Uploaded: 2023-10-30T14:17:24+00:00
Duration: 1 h 16 min 20 s

0:18
hi good morning good often good evening
0:20
hi good morning good often good evening
0:20
hi good morning good often good evening everyone my name is different Simon and
0:22
everyone my name is different Simon and
0:22
everyone my name is different Simon and welcome back to C sharp corner live show
0:24
welcome back to C sharp corner live show
0:24
welcome back to C sharp corner live show once again for this episode we gonna
0:28
once again for this episode we gonna
0:28
once again for this episode we gonna talk about the current and future state
0:30
talk about the current and future state
0:30
talk about the current and future state of voice technology joining us will be
0:33
of voice technology joining us will be
0:33
of voice technology joining us will be Sanjeev and John all the way from United
0:34
Sanjeev and John all the way from United
0:34
Sanjeev and John all the way from United States we all have Alexa and Google
0:38
States we all have Alexa and Google
0:38
States we all have Alexa and Google assistants with us and this this talk is
0:42
assistants with us and this this talk is
0:42
assistants with us and this this talk is very different as we're gonna have a
0:43
very different as we're gonna have a
0:43
very different as we're gonna have a bunch of demos and and a pure go through
0:47
bunch of demos and and a pure go through
0:47
bunch of demos and and a pure go through of the application that if you want to
0:49
of the application that if you want to
0:49
of the application that if you want to start with voice technology and
0:50
start with voice technology and
0:50
start with voice technology and understand how this field is going
0:53
understand how this field is going
0:53
understand how this field is going during this pandemic so without wasting
0:56
during this pandemic so without wasting
0:56
during this pandemic so without wasting a time I would like to pull in Sanjeev
0:58
a time I would like to pull in Sanjeev
0:58
a time I would like to pull in Sanjeev and John our guest for today hi John
1:02
and John our guest for today hi John
1:02
and John our guest for today hi John it's Sanjeev how are you doing good good
1:06
it's Sanjeev how are you doing good good
1:06
it's Sanjeev how are you doing good good yeah we're doing good
1:07
yeah we're doing good
1:07
yeah we're doing good so what why don't I take it
1:11
so what why don't I take it
1:11
so what why don't I take it I'm sorry swathi I'm one of the
1:13
I'm sorry swathi I'm one of the
1:13
I'm sorry swathi I'm one of the cofounders of our start-up whetstone
1:16
cofounders of our start-up whetstone
1:16
cofounders of our start-up whetstone technologies I I'm an X Microsoft guy
1:20
technologies I I'm an X Microsoft guy
1:20
technologies I I'm an X Microsoft guy I've been in the software industry for
1:21
I've been in the software industry for
1:21
I've been in the software industry for about 25 plus years and I was at
1:23
about 25 plus years and I was at
1:23
about 25 plus years and I was at Microsoft I worked on the Windows
1:25
Microsoft I worked on the Windows
1:25
Microsoft I worked on the Windows product team and for Microsoft
1:26
product team and for Microsoft
1:26
product team and for Microsoft consulting services and since this is a
1:29
consulting services and since this is a
1:29
consulting services and since this is a primarily built primarily a developer
1:31
primarily built primarily a developer
1:31
primarily built primarily a developer audience my current coding stack is
1:34
audience my current coding stack is
1:34
audience my current coding stack is c-sharp dotnet porn AWS well that
1:37
c-sharp dotnet porn AWS well that
1:37
c-sharp dotnet porn AWS well that matters without it without community
1:39
matters without it without community
1:39
matters without it without community being right yep and um John II was on
1:46
being right yep and um John II was on
1:46
being right yep and um John II was on the huntress co-founder of Western
1:48
the huntress co-founder of Western
1:48
the huntress co-founder of Western technologies and I also have a similar
1:51
technologies and I also have a similar
1:51
technologies and I also have a similar background I spent 12 years at Microsoft
1:53
background I spent 12 years at Microsoft
1:53
background I spent 12 years at Microsoft in Microsoft consulting services
1:56
in Microsoft consulting services
1:56
in Microsoft consulting services traveling around the country delivering
1:58
traveling around the country delivering
1:58
traveling around the country delivering solutions for clients and now we're
2:01
solutions for clients and now we're
2:01
solutions for clients and now we're focused on delivering voice solutions
2:03
focused on delivering voice solutions
2:03
focused on delivering voice solutions and we're using the stack sans had
2:07
and we're using the stack sans had
2:07
and we're using the stack sans had mentioned c-sharp dotnet core AWS all
2:10
mentioned c-sharp dotnet core AWS all
2:10
mentioned c-sharp dotnet core AWS all built on a service architecture that
2:14
built on a service architecture that
2:14
built on a service architecture that that's great that's great we have some
2:15
that's great that's great we have some
2:15
that's great that's great we have some racing and guys joining us today
2:18
racing and guys joining us today
2:18
racing and guys joining us today why don't you know we're going to anyone
2:20
why don't you know we're going to anyone
2:20
why don't you know we're going to anyone of you just go ahead and give us an idea
2:22
of you just go ahead and give us an idea
2:22
of you just go ahead and give us an idea on what exactly we're going to talk
2:25
on what exactly we're going to talk
2:25
on what exactly we're going to talk today before we move into the slides
2:27
today before we move into the slides
2:27
today before we move into the slides sure so what the way
2:31
sure so what the way
2:31
sure so what the way we've kind of broken the presentation up
2:33
we've kind of broken the presentation up
2:33
we've kind of broken the presentation up give you an idea of what the adoption
2:37
give you an idea of what the adoption
2:37
give you an idea of what the adoption rates you know just what people are
2:38
rates you know just what people are
2:38
rates you know just what people are seeing what are the consumer you know
2:41
seeing what are the consumer you know
2:41
seeing what are the consumer you know how what what does this all mean because
2:43
how what what does this all mean because
2:43
how what what does this all mean because it's not just smart speakers there's so
2:45
it's not just smart speakers there's so
2:45
it's not just smart speakers there's so much more involved in voice technology
2:48
much more involved in voice technology
2:48
much more involved in voice technology we'll give a demo of a voice application
2:52
we'll give a demo of a voice application
2:52
we'll give a demo of a voice application that we're that we have put together
2:55
that we're that we have put together
2:55
that we're that we have put together that's running on that architecture John
2:57
that's running on that architecture John
2:57
that's running on that architecture John just mentioned and then we'll give an
2:59
just mentioned and then we'll give an
2:59
just mentioned and then we'll give an overview of what a voice application is
3:03
overview of what a voice application is
3:03
overview of what a voice application is some of the different challenges you
3:06
some of the different challenges you
3:06
some of the different challenges you know what it takes to write these voice
3:07
know what it takes to write these voice
3:07
know what it takes to write these voice applications and then we're also going
3:10
applications and then we're also going
3:10
applications and then we're also going to talk about some potential future
3:13
to talk about some potential future
3:13
to talk about some potential future state of voice technology and then we'll
3:17
state of voice technology and then we'll
3:17
state of voice technology and then we'll open it up for questions that that's
3:20
open it up for questions that that's
3:20
open it up for questions that that's really sound interesting one note to all
3:23
really sound interesting one note to all
3:23
really sound interesting one note to all the viewers who are watching us right
3:24
the viewers who are watching us right
3:24
the viewers who are watching us right now it may be on YouTube Facebook
3:26
now it may be on YouTube Facebook
3:26
now it may be on YouTube Facebook Twitter or twitch you guys can drop your
3:29
Twitter or twitch you guys can drop your
3:29
Twitter or twitch you guys can drop your questions in the live comment section
3:31
questions in the live comment section
3:31
questions in the live comment section below and John and Sanjeev will be
3:33
below and John and Sanjeev will be
3:33
below and John and Sanjeev will be answering your questions in the life
3:35
answering your questions in the life
3:35
answering your questions in the life episode just after they complete this
3:37
episode just after they complete this
3:37
episode just after they complete this slide and maybe even during the live
3:39
slide and maybe even during the live
3:39
slide and maybe even during the live show so if you have any query to
3:41
show so if you have any query to
3:41
show so if you have any query to whatever they are presenting and if you
3:43
whatever they are presenting and if you
3:43
whatever they are presenting and if you want any guidance for your carrier move
3:46
want any guidance for your carrier move
3:46
want any guidance for your carrier move moving into voice technology you can
3:48
moving into voice technology you can
3:48
moving into voice technology you can just drop comments below and our
3:50
just drop comments below and our
3:50
just drop comments below and our speakers will be answering in the live
3:52
speakers will be answering in the live
3:52
speakers will be answering in the live show having said that I'll leave you and
3:55
show having said that I'll leave you and
3:55
show having said that I'll leave you and John this live show is all yours now
3:57
John this live show is all yours now
3:57
John this live show is all yours now okay I'm gonna pull up your slides and
4:00
okay I'm gonna pull up your slides and
4:00
okay I'm gonna pull up your slides and there you go all right so let's start
4:06
there you go all right so let's start
4:06
there you go all right so let's start with voice technology these days when
4:08
with voice technology these days when
4:08
with voice technology these days when people think about voice technology they
4:11
people think about voice technology they
4:11
people think about voice technology they either think of Alexa or Google
4:13
either think of Alexa or Google
4:13
either think of Alexa or Google assistant and those two those two
4:17
assistant and those two those two
4:17
assistant and those two those two technologies those are assistants they
4:20
technologies those are assistants they
4:20
technologies those are assistants they are voice assistants and that's what's
4:22
are voice assistants and that's what's
4:22
are voice assistants and that's what's driving smart speakers which overall
4:25
driving smart speakers which overall
4:25
driving smart speakers which overall have seen the fastest consumer adoption
4:28
have seen the fastest consumer adoption
4:28
have seen the fastest consumer adoption of technology ever and that includes
4:31
of technology ever and that includes
4:31
of technology ever and that includes cellphones as a January of this year
4:34
cellphones as a January of this year
4:34
cellphones as a January of this year one-third of the adult populations in
4:37
one-third of the adult populations in
4:37
one-third of the adult populations in the United States has installed smart
4:40
the United States has installed smart
4:40
the United States has installed smart speakers in their homes
4:41
speakers in their homes
4:41
speakers in their homes now when you think about smart speakers
4:44
now when you think about smart speakers
4:44
now when you think about smart speakers please note that the numbers that you
4:47
please note that the numbers that you
4:47
please note that the numbers that you know the technology in these numbers
4:49
know the technology in these numbers
4:49
know the technology in these numbers does not include cell phones or other
4:52
does not include cell phones or other
4:52
does not include cell phones or other devices with voice assistance built-in
4:55
devices with voice assistance built-in
4:55
devices with voice assistance built-in such as TVs or cars Google assistant is
4:59
such as TVs or cars Google assistant is
4:59
such as TVs or cars Google assistant is native to Android phones and 2009 in
5:04
native to Android phones and 2009 in
5:04
native to Android phones and 2009 in 2019 at the CES they announced that it
5:08
2019 at the CES they announced that it
5:08
2019 at the CES they announced that it was available on over a billion devices
5:10
was available on over a billion devices
5:10
was available on over a billion devices worldwide this year Google announced
5:14
worldwide this year Google announced
5:14
worldwide this year Google announced that they have 500 million active users
5:17
that they have 500 million active users
5:17
that they have 500 million active users per month spread over 90 countries on
5:20
per month spread over 90 countries on
5:20
per month spread over 90 countries on Google assistant one of the other
5:23
Google assistant one of the other
5:23
Google assistant one of the other interesting announcements they made is
5:24
interesting announcements they made is
5:24
interesting announcements they made is that they're also planning to replace
5:26
that they're also planning to replace
5:26
that they're also planning to replace voice search in Chrome with Google
5:29
voice search in Chrome with Google
5:29
voice search in Chrome with Google Assistant this year so you know when you
5:32
Assistant this year so you know when you
5:32
Assistant this year so you know when you when you start thinking about smart tech
5:34
when you start thinking about smart tech
5:34
when you start thinking about smart tech about voice technology don't just think
5:36
about voice technology don't just think
5:36
about voice technology don't just think at Amazon echo Google home you need to
5:39
at Amazon echo Google home you need to
5:39
at Amazon echo Google home you need to think about the other devices and that
5:43
think about the other devices and that
5:43
think about the other devices and that so now that I've kind of explained you
5:47
so now that I've kind of explained you
5:47
so now that I've kind of explained you know what your possible target audience
5:49
know what your possible target audience
5:49
know what your possible target audience can be what could you do with a voice
5:52
can be what could you do with a voice
5:52
can be what could you do with a voice application so for that I'm going to
5:54
application so for that I'm going to
5:54
application so for that I'm going to give you a short demo here and this is
5:57
give you a short demo here and this is
5:57
give you a short demo here and this is running live on my echo show echo open
6:02
running live on my echo show echo open
6:02
running live on my echo show echo open whetstone technologies welcome back to
6:07
whetstone technologies welcome back to
6:07
whetstone technologies welcome back to whetstone technologies would you like to
6:09
whetstone technologies would you like to
6:09
whetstone technologies would you like to learn about sana bridge or hear more
6:11
learn about sana bridge or hear more
6:11
learn about sana bridge or hear more about whetstone technologies sana bridge
6:16
about whetstone technologies sana bridge
6:16
about whetstone technologies sana bridge sana bridge is our patent-pending
6:18
sana bridge is our patent-pending
6:18
sana bridge is our patent-pending cloud-based platform that enables brands
6:21
cloud-based platform that enables brands
6:21
cloud-based platform that enables brands to monetize Amazon Alexa and Google
6:23
to monetize Amazon Alexa and Google
6:23
to monetize Amazon Alexa and Google assistant voice apps we integrate voice
6:25
assistant voice apps we integrate voice
6:25
assistant voice apps we integrate voice apps with text messaging to deliver
6:27
apps with text messaging to deliver
6:27
apps with text messaging to deliver offers immediately upon user opt-in use
6:30
offers immediately upon user opt-in use
6:30
offers immediately upon user opt-in use cases include product purchase
6:32
cases include product purchase
6:32
cases include product purchase incentives travel and tourism discounts
6:35
incentives travel and tourism discounts
6:35
incentives travel and tourism discounts and offers for professional services
6:37
and offers for professional services
6:37
and offers for professional services want to receive an offer code for a free
6:39
want to receive an offer code for a free
6:39
want to receive an offer code for a free one-hour voice tech consultation in a
6:41
one-hour voice tech consultation in a
6:41
one-hour voice tech consultation in a free white paper about voice technology
6:44
free white paper about voice technology
6:44
free white paper about voice technology sure great I'll need you to provide a
6:48
sure great I'll need you to provide a
6:48
sure great I'll need you to provide a ten digit mobile phone number so I can
6:50
ten digit mobile phone number so I can
6:50
ten digit mobile phone number so I can send you a text message from whetstone
6:52
send you a text message from whetstone
6:52
send you a text message from whetstone technologies using sana grid
6:55
technologies using sana grid
6:55
technologies using sana grid it's six one zero two ninety nine one
6:58
it's six one zero two ninety nine one
6:58
it's six one zero two ninety nine one five seven one I heard you say six one
7:02
five seven one I heard you say six one
7:02
five seven one I heard you say six one zero two nine nine one five seven one if
7:06
zero two nine nine one five seven one if
7:06
zero two nine nine one five seven one if this is correct and you agree to receive
7:08
this is correct and you agree to receive
7:08
this is correct and you agree to receive a text message from whetstone
7:09
a text message from whetstone
7:09
a text message from whetstone technologies using an automatic
7:11
technologies using an automatic
7:11
technologies using an automatic telephone dialing system say yes message
7:14
telephone dialing system say yes message
7:14
telephone dialing system say yes message and data rates may apply yes thank you
7:22
and data rates may apply yes thank you
7:22
and data rates may apply yes thank you I'll send you a text message shortly
7:25
I'll send you a text message shortly
7:25
I'll send you a text message shortly please note that consent to receive text
7:27
please note that consent to receive text
7:27
please note that consent to receive text messages is not a condition of purchase
7:29
messages is not a condition of purchase
7:29
messages is not a condition of purchase goodbye so what I'm sure what we showed
7:36
goodbye so what I'm sure what we showed
7:36
goodbye so what I'm sure what we showed there with that demo is we start with a
7:39
there with that demo is we start with a
7:39
there with that demo is we start with a one ubiquitous device which is a smart
7:42
one ubiquitous device which is a smart
7:42
one ubiquitous device which is a smart speaker that was my echo show and then
7:45
speaker that was my echo show and then
7:45
speaker that was my echo show and then it transitions seamlessly to another
7:48
it transitions seamlessly to another
7:48
it transitions seamlessly to another ubiquitous device which is my cell phone
7:50
ubiquitous device which is my cell phone
7:50
ubiquitous device which is my cell phone and actually for some reason my cell
7:54
and actually for some reason my cell
7:54
and actually for some reason my cell phone didn't ding but I just got the
7:56
phone didn't ding but I just got the
7:56
phone didn't ding but I just got the message delivered on my on my phone and
8:00
message delivered on my on my phone and
8:00
message delivered on my on my phone and you know the thing is is what you can
8:03
you know the thing is is what you can
8:03
you know the thing is is what you can what you can do with this technology
8:05
what you can do with this technology
8:05
what you can do with this technology it's not just you know you don't just
8:08
it's not just you know you don't just
8:08
it's not just you know you don't just think it's something you talk to the way
8:10
think it's something you talk to the way
8:10
think it's something you talk to the way to approach a voice application is what
8:13
to approach a voice application is what
8:13
to approach a voice application is what can you do to make other things happen
8:16
can you do to make other things happen
8:16
can you do to make other things happen and that you know whether you're turning
8:18
and that you know whether you're turning
8:18
and that you know whether you're turning on lights whether you're starting your
8:21
on lights whether you're starting your
8:21
on lights whether you're starting your xbox there's you know you know what can
8:24
xbox there's you know you know what can
8:24
xbox there's you know you know what can you do to leverage you know to leverage
8:27
you do to leverage you know to leverage
8:27
you do to leverage you know to leverage and act as another mode in interacting
8:30
and act as another mode in interacting
8:30
and act as another mode in interacting with other types of technology so just
8:34
with other types of technology so just
8:34
with other types of technology so just to give you a note that particular
8:36
to give you a note that particular
8:36
to give you a note that particular application that's that was running on
8:39
application that's that was running on
8:39
application that's that was running on Alexa that's called an Alexa skill that
8:41
Alexa that's called an Alexa skill that
8:41
Alexa that's called an Alexa skill that is available and running live right now
8:44
is available and running live right now
8:44
is available and running live right now it's currently in the United States only
8:46
it's currently in the United States only
8:46
it's currently in the United States only just because we're this one who's
8:48
just because we're this one who's
8:48
just because we're this one who's working with text messaging and now to
8:50
working with text messaging and now to
8:50
working with text messaging and now to just give you an idea of how voice
8:52
just give you an idea of how voice
8:52
just give you an idea of how voice applications work and some of the what
8:55
applications work and some of the what
8:55
applications work and some of the what it takes to develop them I'm going to
8:57
it takes to develop them I'm going to
8:57
it takes to develop them I'm going to turn everything over to Jon Snow Oh
9:04
turn everything over to Jon Snow Oh
9:04
turn everything over to Jon Snow Oh so what you just saw was a
9:07
so what you just saw was a
9:07
so what you just saw was a demonstration of Alexis skill we
9:10
demonstration of Alexis skill we
9:10
demonstration of Alexis skill we generally use the term voice application
9:13
generally use the term voice application
9:13
generally use the term voice application to mean the broader umbrella of voice
9:17
to mean the broader umbrella of voice
9:17
to mean the broader umbrella of voice experiences Alexa calls them skills
9:19
experiences Alexa calls them skills
9:19
experiences Alexa calls them skills Google caused them actions sense and
9:23
Google caused them actions sense and
9:23
Google caused them actions sense and Bixby calls them capsules but whatever
9:25
Bixby calls them capsules but whatever
9:25
Bixby calls them capsules but whatever you call them they they are voice
9:28
you call them they they are voice
9:28
you call them they they are voice applications and follow a similar model
9:32
applications and follow a similar model
9:32
applications and follow a similar model of the natural language they follow a
9:36
of the natural language they follow a
9:36
of the natural language they follow a natural language model that you must
9:38
natural language model that you must
9:38
natural language model that you must configure before any of these
9:40
configure before any of these
9:40
configure before any of these applications go live that NLP were
9:45
applications go live that NLP were
9:45
applications go live that NLP were rather NLM is comprised of the
9:49
rather NLM is comprised of the
9:49
rather NLM is comprised of the invocation name you just heard songs use
9:52
invocation name you just heard songs use
9:52
invocation name you just heard songs use Western technologies to open the skill
9:55
Western technologies to open the skill
9:55
Western technologies to open the skill it's the unique identifier that can be
10:02
it's the unique identifier that can be
10:02
it's the unique identifier that can be used to invoke one of these voice
10:03
used to invoke one of these voice
10:03
used to invoke one of these voice applications on a smart speaker there's
10:07
applications on a smart speaker there's
10:07
applications on a smart speaker there's a number of design considerations that
10:09
a number of design considerations that
10:09
a number of design considerations that goes into selecting what that invocation
10:12
goes into selecting what that invocation
10:12
goes into selecting what that invocation name is you want it to be unique
10:16
name is you want it to be unique
10:16
name is you want it to be unique relatively easy to pronounce you can be
10:22
relatively easy to pronounce you can be
10:22
relatively easy to pronounce you can be mindful of brand names for example if
10:24
mindful of brand names for example if
10:24
mindful of brand names for example if you have a brand name that's a little
10:26
you have a brand name that's a little
10:26
you have a brand name that's a little hard to pronounce for a business you may
10:29
hard to pronounce for a business you may
10:29
hard to pronounce for a business you may want to shy away from using that use
10:31
want to shy away from using that use
10:31
want to shy away from using that use something a little bit more ubiquitous
10:34
something a little bit more ubiquitous
10:34
something a little bit more ubiquitous once the voice application is launched
10:37
once the voice application is launched
10:37
once the voice application is launched it's a question answer scenario you know
10:41
it's a question answer scenario you know
10:41
it's a question answer scenario you know you you invoke the voice application it
10:45
you you invoke the voice application it
10:45
you you invoke the voice application it gives you a brief introduction and then
10:47
gives you a brief introduction and then
10:47
gives you a brief introduction and then prompts you with what you can do next
10:52
prompts you with what you can do next
10:52
prompts you with what you can do next the next statement you make is an intent
10:56
the next statement you make is an intent
10:56
the next statement you make is an intent that's effectively a voice command
10:58
that's effectively a voice command
10:58
that's effectively a voice command you've heard Songz say yes when asked if
11:03
you've heard Songz say yes when asked if
11:03
you've heard Songz say yes when asked if he wanted to receive a a white paper
11:09
he wanted to receive a a white paper
11:09
he wanted to receive a a white paper from Western technologies that yes rolls
11:13
from Western technologies that yes rolls
11:13
from Western technologies that yes rolls up to a yes intent that your your voice
11:18
up to a yes intent that your your voice
11:18
up to a yes intent that your your voice application processes
11:22
other commands couldn't could include no
11:26
other commands couldn't could include no
11:26
other commands couldn't could include no book of flights booked a hotel usually
11:33
book of flights booked a hotel usually
11:33
book of flights booked a hotel usually these intents are you know they can be
11:38
these intents are you know they can be
11:38
these intents are you know they can be expressed multiple ways you can say yes
11:41
expressed multiple ways you can say yes
11:41
expressed multiple ways you can say yes you can say yeah you can say okay each
11:44
you can say yeah you can say okay each
11:44
you can say yeah you can say okay each of those different phrases is an
11:46
of those different phrases is an
11:46
of those different phrases is an utterance that rolls up to an intent so
11:49
utterance that rolls up to an intent so
11:49
utterance that rolls up to an intent so as a skill developer as a voice
11:51
as a skill developer as a voice
11:51
as a skill developer as a voice application developer the inbound
11:53
application developer the inbound
11:53
application developer the inbound command that you get from the users
11:56
command that you get from the users
11:56
command that you get from the users input into the device is okay they ask
12:00
input into the device is okay they ask
12:00
input into the device is okay they ask for a yes intent but you don't
12:02
for a yes intent but you don't
12:02
for a yes intent but you don't necessarily necessarily know exactly
12:04
necessarily necessarily know exactly
12:04
necessarily necessarily know exactly what they said they could have said sure
12:06
what they said they could have said sure
12:06
what they said they could have said sure okay go ahead you know that you
12:11
okay go ahead you know that you
12:11
okay go ahead you know that you generally don't get that fine-grained
12:14
generally don't get that fine-grained
12:14
generally don't get that fine-grained information in in the inbound requests
12:17
information in in the inbound requests
12:17
information in in the inbound requests but you know the intent by itself is not
12:23
but you know the intent by itself is not
12:23
but you know the intent by itself is not sufficient to complete the user
12:25
sufficient to complete the user
12:25
sufficient to complete the user experience you heard song provide a
12:27
experience you heard song provide a
12:27
experience you heard song provide a phone number that phone number is
12:29
phone number that phone number is
12:29
phone number that phone number is provided by a slot which is essentially
12:32
provided by a slot which is essentially
12:32
provided by a slot which is essentially a variable so you would have a phone
12:36
a variable so you would have a phone
12:36
a variable so you would have a phone number intent the utterance would be
12:39
number intent the utterance would be
12:39
number intent the utterance would be something like my phone number is my
12:42
something like my phone number is my
12:42
something like my phone number is my number is and what follows would be your
12:47
number is and what follows would be your
12:47
number is and what follows would be your spoken phone number that would come into
12:49
spoken phone number that would come into
12:49
spoken phone number that would come into your voice application as a slot as a
12:52
your voice application as a slot as a
12:52
your voice application as a slot as a variable slots can have synonyms in the
12:59
variable slots can have synonyms in the
12:59
variable slots can have synonyms in the book a book a flight scenario you could
13:04
book a book a flight scenario you could
13:04
book a book a flight scenario you could say what city you're going to write I'm
13:06
say what city you're going to write I'm
13:06
say what city you're going to write I'm traveling to you know Philadelphia well
13:09
traveling to you know Philadelphia well
13:09
traveling to you know Philadelphia well I happen to be based out of Philadelphia
13:11
I happen to be based out of Philadelphia
13:11
I happen to be based out of Philadelphia and it's common for people in the
13:15
and it's common for people in the
13:15
and it's common for people in the Northeast and across the u.s. to refer
13:18
Northeast and across the u.s. to refer
13:18
Northeast and across the u.s. to refer to Philadelphia as Philly so you may
13:20
to Philadelphia as Philly so you may
13:20
to Philadelphia as Philly so you may have a synonym attached to the slot
13:24
have a synonym attached to the slot
13:24
have a synonym attached to the slot variable destination and the person may
13:27
variable destination and the person may
13:27
variable destination and the person may say Philly or Philadelphia but you'll
13:30
say Philly or Philadelphia but you'll
13:30
say Philly or Philadelphia but you'll receive Philadelphia as the voice up
13:34
receive Philadelphia as the voice up
13:34
receive Philadelphia as the voice up legate
13:35
legate
13:35
legate bellever when you do decide to lay out
13:40
bellever when you do decide to lay out
13:40
bellever when you do decide to lay out your natural language model it's
13:43
your natural language model it's
13:43
your natural language model it's important to consider the flow of the
13:45
important to consider the flow of the
13:45
important to consider the flow of the application whenever you you know write
13:48
application whenever you you know write
13:48
application whenever you you know write a user interface it's common for
13:50
a user interface it's common for
13:50
a user interface it's common for developers to start with wireframes it's
13:52
developers to start with wireframes it's
13:52
developers to start with wireframes it's usually the user experience that's
13:55
usually the user experience that's
13:55
usually the user experience that's discussed and ironed out before the
13:57
discussed and ironed out before the
13:57
discussed and ironed out before the first line of code is written the same
14:00
first line of code is written the same
14:00
first line of code is written the same holds when you're designing a voice
14:01
holds when you're designing a voice
14:01
holds when you're designing a voice application you really should lay out a
14:04
application you really should lay out a
14:04
application you really should lay out a script and a script is experience which
14:08
script and a script is experience which
14:08
script and a script is experience which will inform what intense you need what
14:12
will inform what intense you need what
14:12
will inform what intense you need what prompts you need and it's helpful when
14:15
prompts you need and it's helpful when
14:15
prompts you need and it's helpful when you're designing this experience to sit
14:17
you're designing this experience to sit
14:17
you're designing this experience to sit down with somebody that is unfamiliar
14:21
down with somebody that is unfamiliar
14:21
down with somebody that is unfamiliar with what you're trying to do and act
14:24
with what you're trying to do and act
14:24
with what you're trying to do and act out that script with them to see if the
14:27
out that script with them to see if the
14:27
out that script with them to see if the prompts that you use elicits the
14:30
prompts that you use elicits the
14:30
prompts that you use elicits the responses that you expect and the
14:32
responses that you expect and the
14:32
responses that you expect and the intents you expect that design practice
14:37
intents you expect that design practice
14:37
intents you expect that design practice is called voice ux we're all we're
14:40
is called voice ux we're all we're
14:40
is called voice ux we're all we're generally familiar familiar with user
14:43
generally familiar familiar with user
14:43
generally familiar familiar with user experience design and UX this is voice
14:47
experience design and UX this is voice
14:47
experience design and UX this is voice ux v ux if you search for it online I'm
14:50
ux v ux if you search for it online I'm
14:50
ux v ux if you search for it online I'm sure you'll you'll find it it may also
14:52
sure you'll you'll find it it may also
14:52
sure you'll you'll find it it may also be referred to as voice user interface
14:55
be referred to as voice user interface
14:55
be referred to as voice user interface Bui so again one of the first things you
15:00
Bui so again one of the first things you
15:00
Bui so again one of the first things you should do before you write your voice
15:02
should do before you write your voice
15:02
should do before you write your voice application as you know script it out
15:04
application as you know script it out
15:04
application as you know script it out that will inform your natural language
15:06
that will inform your natural language
15:06
that will inform your natural language model and work with somebody to test out
15:09
model and work with somebody to test out
15:09
model and work with somebody to test out that script so now we're onto the
15:18
that script so now we're onto the
15:18
that script so now we're onto the response side of the house so when Alexa
15:21
response side of the house so when Alexa
15:21
response side of the house so when Alexa gets a request
15:21
gets a request
15:22
gets a request Alexa sends a response the way those
15:24
Alexa sends a response the way those
15:24
Alexa sends a response the way those responses are sent is either you know
15:27
responses are sent is either you know
15:27
responses are sent is either you know plain text but that could be read kind
15:32
plain text but that could be read kind
15:32
plain text but that could be read kind of flatly you know Alexa you want to
15:35
of flatly you know Alexa you want to
15:35
of flatly you know Alexa you want to give a let's us some personality that
15:37
give a let's us some personality that
15:37
give a let's us some personality that makes your skill stand out so one way to
15:39
makes your skill stand out so one way to
15:39
makes your skill stand out so one way to do that is to return an SS ml response
15:43
do that is to return an SS ml response
15:43
do that is to return an SS ml response that's speech synthesis markup language
15:47
that's speech synthesis markup language
15:47
that's speech synthesis markup language say that ten times
15:49
say that ten times
15:49
say that ten times it's a format of XML and it's an open
15:55
it's a format of XML and it's an open
15:55
it's a format of XML and it's an open standard and with most open standards
15:59
standard and with most open standards
15:59
standard and with most open standards you'll find that different platforms
16:01
you'll find that different platforms
16:01
you'll find that different platforms have slightly different implementations
16:04
have slightly different implementations
16:04
have slightly different implementations of its both find on Alexa is different
16:11
of its both find on Alexa is different
16:11
of its both find on Alexa is different than what you'll find on Google in terms
16:12
than what you'll find on Google in terms
16:12
than what you'll find on Google in terms of the different tags that SSM Alice
16:15
of the different tags that SSM Alice
16:15
of the different tags that SSM Alice opposes support and some platforms have
16:18
opposes support and some platforms have
16:18
opposes support and some platforms have custom tags that they've they've added
16:21
custom tags that they've they've added
16:21
custom tags that they've they've added there is an effort to normalize this
16:25
there is an effort to normalize this
16:25
there is an effort to normalize this which is called speech markup that's a I
16:30
which is called speech markup that's a I
16:30
which is called speech markup that's a I said just you google that if you're
16:34
said just you google that if you're
16:34
said just you google that if you're interested in learning more about s s ml
16:35
interested in learning more about s s ml
16:35
interested in learning more about s s ml its attempts to normalize the expression
16:40
its attempts to normalize the expression
16:40
its attempts to normalize the expression of s s ml across multiple platforms so
16:43
of s s ml across multiple platforms so
16:43
of s s ml across multiple platforms so in these three examples they'll show you
16:46
in these three examples they'll show you
16:46
in these three examples they'll show you how you can change the speed of the
16:51
how you can change the speed of the
16:51
how you can change the speed of the return speech using prosody so Sandra P
16:55
return speech using prosody so Sandra P
16:55
return speech using prosody so Sandra P will a nickel swing would have been
17:01
will a nickel swing would have been
17:01
will a nickel swing would have been ILI's I can speak with my normal pitch
17:04
ILI's I can speak with my normal pitch
17:04
ILI's I can speak with my normal pitch but also with a much higher pitch
17:06
but also with a much higher pitch
17:06
but also with a much higher pitch investor and also with the lower pitch
17:08
investor and also with the lower pitch
17:08
investor and also with the lower pitch and slowly right so you just saw how you
17:12
and slowly right so you just saw how you
17:12
and slowly right so you just saw how you can manipulate Alexa Alex's voice this
17:18
can manipulate Alexa Alex's voice this
17:18
can manipulate Alexa Alex's voice this becomes handy when you're talking when
17:20
becomes handy when you're talking when
17:20
becomes handy when you're talking when you're considering accessibility
17:22
you're considering accessibility
17:22
you're considering accessibility scenarios where if you're designing a
17:24
scenarios where if you're designing a
17:24
scenarios where if you're designing a skill for somebody that may be a bit
17:27
skill for somebody that may be a bit
17:27
skill for somebody that may be a bit hard of hearing you can increase the
17:29
hard of hearing you can increase the
17:29
hard of hearing you can increase the volume if were decrease the rate of
17:33
volume if were decrease the rate of
17:33
volume if were decrease the rate of speech so again consider your target
17:36
speech so again consider your target
17:36
speech so again consider your target audience when you're putting your s ml
17:39
audience when you're putting your s ml
17:39
audience when you're putting your s ml responses together you can also override
17:42
responses together you can also override
17:42
responses together you can also override the pronunciation of various words again
17:46
the pronunciation of various words again
17:46
the pronunciation of various words again with brand names you'll were last names
17:50
with brand names you'll were last names
17:50
with brand names you'll were last names you'll find that the smart speakers
17:52
you'll find that the smart speakers
17:53
you'll find that the smart speakers generally don't pronounce them quite
17:54
generally don't pronounce them quite
17:54
generally don't pronounce them quite right well you can modify that and again
17:59
right well you can modify that and again
17:59
right well you can modify that and again you know since I am out of Philadelphia
18:01
you know since I am out of Philadelphia
18:01
you know since I am out of Philadelphia Arlo
18:02
Arlo
18:02
Arlo our team is called the Philadelphia
18:04
our team is called the Philadelphia
18:04
our team is called the Philadelphia Eagles but the local field off your
18:09
Eagles but the local field off your
18:09
Eagles but the local field off your accent is a bit odd so sometimes you
18:12
accent is a bit odd so sometimes you
18:12
accent is a bit odd so sometimes you might hear somebody refer to Eagles as
18:14
might hear somebody refer to Eagles as
18:14
might hear somebody refer to Eagles as Ebel's and so I went to design an
18:17
Ebel's and so I went to design an
18:17
Ebel's and so I went to design an experience that was a little more closer
18:19
experience that was a little more closer
18:19
experience that was a little more closer to how Philadelphians speak and so you
18:23
to how Philadelphians speak and so you
18:23
to how Philadelphians speak and so you can hear this example a new cool swing
18:31
can hear this example a new cool swing
18:31
can hear this example a new cool swing would have been Millie's do that one
18:34
would have been Millie's do that one
18:34
would have been Millie's do that one more time that went by fairly quickly a
18:38
more time that went by fairly quickly a
18:38
more time that went by fairly quickly a nickel swing would have been Millie's
18:40
nickel swing would have been Millie's
18:40
nickel swing would have been Millie's right so you rather than Eagles you
18:44
right so you rather than Eagles you
18:44
right so you rather than Eagles you heard Eagles and the the word nice is
18:48
heard Eagles and the the word nice is
18:48
heard Eagles and the the word nice is sometimes pronounced noise with that
18:51
sometimes pronounced noise with that
18:51
sometimes pronounced noise with that with that long oh and then finally you
18:56
with that long oh and then finally you
18:57
with that long oh and then finally you can have short relatively short audio
19:01
can have short relatively short audio
19:01
can have short relatively short audio files embedded in your response that's
19:05
files embedded in your response that's
19:05
files embedded in your response that's useful if you're you know creating a
19:07
useful if you're you know creating a
19:07
useful if you're you know creating a game or if you want to create a branded
19:10
game or if you want to create a branded
19:10
game or if you want to create a branded experience with a jingle or sometimes
19:15
experience with a jingle or sometimes
19:15
experience with a jingle or sometimes it's referred to as Sonic branding when
19:17
it's referred to as Sonic branding when
19:17
it's referred to as Sonic branding when you want to associate a sound with a
19:19
you want to associate a sound with a
19:19
you want to associate a sound with a particular brand or business so Sandra
19:23
particular brand or business so Sandra
19:23
particular brand or business so Sandra whew well there's a troll in your way
19:34
whew well there's a troll in your way
19:34
whew well there's a troll in your way right so this has two interesting
19:38
right so this has two interesting
19:38
right so this has two interesting examples of embedding the mp3 and also
19:42
examples of embedding the mp3 and also
19:42
examples of embedding the mp3 and also changing the voice that is coming back
19:47
changing the voice that is coming back
19:47
changing the voice that is coming back Alexa supports Amazon Kali which
19:50
Alexa supports Amazon Kali which
19:50
Alexa supports Amazon Kali which includes a you know a few dozen
19:53
includes a you know a few dozen
19:53
includes a you know a few dozen different voices that are designed for
19:55
different voices that are designed for
19:55
different voices that are designed for different dialects Emma happens to be a
19:58
different dialects Emma happens to be a
19:58
different dialects Emma happens to be a uk-based voice and this voice name tag
20:03
uk-based voice and this voice name tag
20:03
uk-based voice and this voice name tag that you see there is specific to Amazon
20:08
that you see there is specific to Amazon
20:08
that you see there is specific to Amazon you wouldn't be able to go to say Google
20:11
you wouldn't be able to go to say Google
20:11
you wouldn't be able to go to say Google Google action and use the same tag it's
20:15
Google action and use the same tag it's
20:15
Google action and use the same tag it's very close
20:15
very close
20:15
very close tied to Amazon so let's say you wanted
20:22
tied to Amazon so let's say you wanted
20:22
tied to Amazon so let's say you wanted to create an Alexa skill there's a
20:26
to create an Alexa skill there's a
20:26
to create an Alexa skill there's a developer console developer amazon.com
20:29
developer console developer amazon.com
20:29
developer console developer amazon.com which lets you stand up and lets a skill
20:32
which lets you stand up and lets a skill
20:32
which lets you stand up and lets a skill to find your natural language model and
20:34
to find your natural language model and
20:34
to find your natural language model and wire it to a back-end that back-end
20:38
wire it to a back-end that back-end
20:38
wire it to a back-end that back-end could be supported by serverless lambda
20:42
could be supported by serverless lambda
20:42
could be supported by serverless lambda function if you're familiar with Azure
20:43
function if you're familiar with Azure
20:43
function if you're familiar with Azure it's Amazon's answer to a junction or
20:48
it's Amazon's answer to a junction or
20:48
it's Amazon's answer to a junction or you can have it go to a rest endpoint
20:51
you can have it go to a rest endpoint
20:51
you can have it go to a rest endpoint which you can host of course in your
20:53
which you can host of course in your
20:53
which you can host of course in your cloud provider of choice or on Prem at
20:56
cloud provider of choice or on Prem at
20:56
cloud provider of choice or on Prem at that point of course it doesn't matter
20:58
that point of course it doesn't matter
20:58
that point of course it doesn't matter it's your your decision where you wish
21:00
it's your your decision where you wish
21:00
it's your your decision where you wish to place it the the smart speaker call
21:08
to place it the the smart speaker call
21:08
to place it the the smart speaker call it is made to Amazon hosted services in
21:12
it is made to Amazon hosted services in
21:12
it is made to Amazon hosted services in AWS and then an Alexa request that's a
21:17
AWS and then an Alexa request that's a
21:17
AWS and then an Alexa request that's a JSON message that contains the intense
21:19
JSON message that contains the intense
21:19
JSON message that contains the intense lot values user identifier is all sent
21:23
lot values user identifier is all sent
21:23
lot values user identifier is all sent to your REST API or lambda and you
21:26
to your REST API or lambda and you
21:26
to your REST API or lambda and you return an Alexa response and your Amazon
21:33
return an Alexa response and your Amazon
21:33
return an Alexa response and your Amazon developer console is used to create kind
21:36
developer console is used to create kind
21:36
developer console is used to create kind of a developer instance where you can
21:39
of a developer instance where you can
21:39
of a developer instance where you can test it locally there is a test harness
21:41
test it locally there is a test harness
21:41
test it locally there is a test harness you can use before it goes live and
21:43
you can use before it goes live and
21:43
you can use before it goes live and that's where you can determine and wire
21:47
that's where you can determine and wire
21:47
that's where you can determine and wire up a rest endpoint or lambda endpoint so
21:52
up a rest endpoint or lambda endpoint so
21:52
up a rest endpoint or lambda endpoint so we've launched the Google actions so
21:55
we've launched the Google actions so
21:55
we've launched the Google actions so there are some similarities here just as
21:58
there are some similarities here just as
21:58
there are some similarities here just as you have the Alexa Developer console you
22:01
you have the Alexa Developer console you
22:01
you have the Alexa Developer console you have a Google actions console where
22:04
have a Google actions console where
22:04
have a Google actions console where you'll create your your Google action as
22:08
you'll create your your Google action as
22:08
you'll create your your Google action as well as the metadata that goes around
22:09
well as the metadata that goes around
22:09
well as the metadata that goes around the Google action like the invocation
22:11
the Google action like the invocation
22:11
the Google action like the invocation name when you publish either skill or
22:16
name when you publish either skill or
22:16
name when you publish either skill or Google action there's a description text
22:20
Google action there's a description text
22:20
Google action there's a description text you can set there's a an icon that you
22:24
you can set there's a an icon that you
22:24
you can set there's a an icon that you associate with these skills and actions
22:27
associate with these skills and actions
22:27
associate with these skills and actions that shows up in the
22:29
that shows up in the
22:29
that shows up in the action console and just as you can
22:35
action console and just as you can
22:35
action console and just as you can configure a lambda or rest endpoint in
22:38
configure a lambda or rest endpoint in
22:38
configure a lambda or rest endpoint in and looks a skill you could use a Google
22:41
and looks a skill you could use a Google
22:41
and looks a skill you could use a Google cloud function or a rest endpoint and
22:44
cloud function or a rest endpoint and
22:44
cloud function or a rest endpoint and your there's a little bit of a
22:48
your there's a little bit of a
22:48
your there's a little bit of a difference between how the NLP model is
22:50
difference between how the NLP model is
22:50
difference between how the NLP model is defined in this user experience in this
22:54
defined in this user experience in this
22:54
defined in this user experience in this developer experience versus house
22:55
developer experience versus house
22:55
developer experience versus house defined in Alexa the action console is a
23:00
defined in Alexa the action console is a
23:00
defined in Alexa the action console is a different interface than the dialogue
23:02
different interface than the dialogue
23:02
different interface than the dialogue flow console so you can find your NLP
23:04
flow console so you can find your NLP
23:04
flow console so you can find your NLP model over Annoying model in dialogue
23:07
model over Annoying model in dialogue
23:07
model over Annoying model in dialogue flow and you define your action metadata
23:11
flow and you define your action metadata
23:11
flow and you define your action metadata and you run your test harness and the
23:14
and you run your test harness and the
23:14
and you run your test harness and the action console it's a it takes a bit of
23:16
action console it's a it takes a bit of
23:16
action console it's a it takes a bit of getting used to and there is an
23:19
getting used to and there is an
23:19
getting used to and there is an important distinction between Google
23:21
important distinction between Google
23:21
important distinction between Google Apps and Google actions and your Alexis
23:24
Apps and Google actions and your Alexis
23:24
Apps and Google actions and your Alexis skill Google gives you a little bit more
23:27
skill Google gives you a little bit more
23:27
skill Google gives you a little bit more information on the inbound requests on
23:30
information on the inbound requests on
23:30
information on the inbound requests on the dialogue flow requests that that's
23:31
the dialogue flow requests that that's
23:32
the dialogue flow requests that that's coming into your skill it includes the
23:35
coming into your skill it includes the
23:35
coming into your skill it includes the this what the Google device thinks it
23:41
this what the Google device thinks it
23:41
this what the Google device thinks it heard the person say so on Alexa if
23:45
heard the person say so on Alexa if
23:45
heard the person say so on Alexa if somebody said yes yeah okay sure go
23:49
somebody said yes yeah okay sure go
23:49
somebody said yes yeah okay sure go ahead for a yes intent the only thing
23:52
ahead for a yes intent the only thing
23:52
ahead for a yes intent the only thing you'll see as a developer is the yes
23:53
you'll see as a developer is the yes
23:53
you'll see as a developer is the yes intent and bound on the request with
23:57
intent and bound on the request with
23:57
intent and bound on the request with Google actions to be a little more
23:59
Google actions to be a little more
23:59
Google actions to be a little more information you get what Google thinks
24:02
information you get what Google thinks
24:02
information you get what Google thinks the person said so you'll get the yes
24:05
the person said so you'll get the yes
24:05
the person said so you'll get the yes intent but you'll also get the
24:08
intent but you'll also get the
24:08
intent but you'll also get the speech-to-text transcripts of what the
24:11
speech-to-text transcripts of what the
24:11
speech-to-text transcripts of what the person said that's incredibly useful if
24:14
person said that's incredibly useful if
24:14
person said that's incredibly useful if you have a something that the person
24:21
you have a something that the person
24:21
you have a something that the person said that doesn't map to an intent that
24:24
said that doesn't map to an intent that
24:24
said that doesn't map to an intent that you had already defined let's say you
24:26
you had already defined let's say you
24:26
you had already defined let's say you expected the person to say yes no but
24:30
expected the person to say yes no but
24:30
expected the person to say yes no but they said maybe and you didn't account
24:32
they said maybe and you didn't account
24:32
they said maybe and you didn't account for that on Alexa you would get a
24:35
for that on Alexa you would get a
24:35
for that on Alexa you would get a fallback intent and that's also you
24:37
fallback intent and that's also you
24:37
fallback intent and that's also you wouldn't get what the person said on
24:40
wouldn't get what the person said on
24:40
wouldn't get what the person said on Google you would get
24:43
Google you would get
24:43
Google you would get a the equivalent of a fallback intent
24:45
a the equivalent of a fallback intent
24:45
a the equivalent of a fallback intent but you also get what the person said so
24:48
but you also get what the person said so
24:48
but you also get what the person said so you can you'll get that maybe and you
24:53
you can you'll get that maybe and you
24:53
you can you'll get that maybe and you can use that to inform update updates to
24:55
can use that to inform update updates to
24:55
can use that to inform update updates to your natural language model so you're
24:57
your natural language model so you're
24:57
your natural language model so you're probably not going to get it right the
24:59
probably not going to get it right the
24:59
probably not going to get it right the very first time you publish your Google
25:03
very first time you publish your Google
25:03
very first time you publish your Google action or Alexa skill I would suggest
25:07
action or Alexa skill I would suggest
25:07
action or Alexa skill I would suggest that you publish on both platforms
25:10
that you publish on both platforms
25:10
that you publish on both platforms because what you get on Google is
25:14
because what you get on Google is
25:14
because what you get on Google is probably a similar experience that
25:16
probably a similar experience that
25:16
probably a similar experience that people are having on Alexa but you can
25:18
people are having on Alexa but you can
25:18
people are having on Alexa but you can use the the recorded speech that Google
25:24
use the the recorded speech that Google
25:24
use the the recorded speech that Google gives you the recorded utterance that
25:27
gives you the recorded utterance that
25:27
gives you the recorded utterance that Google gives you that will help inform
25:29
Google gives you that will help inform
25:29
Google gives you that will help inform how you change your natural language
25:31
how you change your natural language
25:31
how you change your natural language model and as from a back-end standpoint
25:37
model and as from a back-end standpoint
25:37
model and as from a back-end standpoint Alexa will send you to two requests well
25:43
Alexa will send you to two requests well
25:43
Alexa will send you to two requests well it'll send you an initial request and
25:46
it'll send you an initial request and
25:46
it'll send you an initial request and then if you don't which if your REST API
25:48
then if you don't which if your REST API
25:48
then if you don't which if your REST API you were your lambda does not respond
25:51
you were your lambda does not respond
25:51
you were your lambda does not respond within a couple of seconds it'll send
25:52
within a couple of seconds it'll send
25:52
within a couple of seconds it'll send you a second request and it'll expect
25:56
you a second request and it'll expect
25:56
you a second request and it'll expect you to respond within eight seconds if
25:58
you to respond within eight seconds if
25:58
you to respond within eight seconds if it doesn't then the skill of ends and
26:01
it doesn't then the skill of ends and
26:01
it doesn't then the skill of ends and the user is removed from your skill
26:06
the user is removed from your skill
26:06
the user is removed from your skill Google action will send you one and only
26:09
Google action will send you one and only
26:09
Google action will send you one and only one request and expect a six second SLA
26:12
one request and expect a six second SLA
26:12
one request and expect a six second SLA response time it will also send a health
26:16
response time it will also send a health
26:16
response time it will also send a health check request coming in and that'll come
26:21
check request coming in and that'll come
26:21
check request coming in and that'll come in about every 10 minutes it's like a
26:23
in about every 10 minutes it's like a
26:23
in about every 10 minutes it's like a ping request to make sure that your
26:25
ping request to make sure that your
26:25
ping request to make sure that your endpoint is still responsive we found
26:29
endpoint is still responsive we found
26:29
endpoint is still responsive we found that Google's documentation is not quite
26:32
that Google's documentation is not quite
26:32
that Google's documentation is not quite as crystal clear as Amazon's
26:35
as crystal clear as Amazon's
26:35
as crystal clear as Amazon's documentation is so if you are looking
26:37
documentation is so if you are looking
26:37
documentation is so if you are looking to publish a Google action make sure you
26:40
to publish a Google action make sure you
26:40
to publish a Google action make sure you search the documentation for a health
26:43
search the documentation for a health
26:43
search the documentation for a health check because we had an experience when
26:46
check because we had an experience when
26:46
check because we had an experience when the first time we published a Google
26:47
the first time we published a Google
26:47
the first time we published a Google action we missed the documentation on
26:50
action we missed the documentation on
26:50
action we missed the documentation on the health check and the skill went lot
26:52
the health check and the skill went lot
26:52
the health check and the skill went lot that Google action went live
26:54
that Google action went live
26:54
that Google action went live surprisingly without
26:56
surprisingly without
26:56
surprisingly without you know the health track being
26:59
you know the health track being
26:59
you know the health track being monitored and being responded to and
27:01
monitored and being responded to and
27:01
monitored and being responded to and then within 10 minutes of it going live
27:03
then within 10 minutes of it going live
27:03
then within 10 minutes of it going live we started getting getting notifications
27:06
we started getting getting notifications
27:06
we started getting getting notifications from Google saying your action is
27:10
from Google saying your action is
27:10
from Google saying your action is non-responsive it's about to be taken
27:11
non-responsive it's about to be taken
27:11
non-responsive it's about to be taken down and it took us going through some
27:13
down and it took us going through some
27:13
down and it took us going through some logs to find out exactly what was going
27:15
logs to find out exactly what was going
27:15
logs to find out exactly what was going on so don't miss that nuance it's we
27:19
on so don't miss that nuance it's we
27:19
on so don't miss that nuance it's we were very surprised that we were able to
27:21
were very surprised that we were able to
27:21
were very surprised that we were able to put a Google action into production
27:23
put a Google action into production
27:23
put a Google action into production without the help check in place okay you
27:28
without the help check in place okay you
27:28
without the help check in place okay you know I'm gonna close out one last one
27:31
know I'm gonna close out one last one
27:31
know I'm gonna close out one last one last voice assistant which is Samsung's
27:34
last voice assistant which is Samsung's
27:34
last voice assistant which is Samsung's big speed
27:35
big speed
27:35
big speed that one's relatively new having come
27:38
that one's relatively new having come
27:38
that one's relatively new having come out you know really there
27:40
out you know really there
27:40
out you know really there Samsung really started pushing what they
27:42
Samsung really started pushing what they
27:42
Samsung really started pushing what they call bixby 2.0 sometime around fall
27:45
call bixby 2.0 sometime around fall
27:45
call bixby 2.0 sometime around fall summer to fall of last year with
27:49
summer to fall of last year with
27:49
summer to fall of last year with Samsung's Bigsby what you end up
27:51
Samsung's Bigsby what you end up
27:51
Samsung's Bigsby what you end up developing is what's called a Bigsby
27:53
developing is what's called a Bigsby
27:53
developing is what's called a Bigsby capsule and Bigsby 2.0 is available on
27:58
capsule and Bigsby 2.0 is available on
27:58
capsule and Bigsby 2.0 is available on samsung smartphones and TVs and
28:00
samsung smartphones and TVs and
28:01
samsung smartphones and TVs and eventually it's going to be coming to
28:02
eventually it's going to be coming to
28:02
eventually it's going to be coming to appliances now Samsung has taken a very
28:07
appliances now Samsung has taken a very
28:07
appliances now Samsung has taken a very opinionated approach that's very
28:09
opinionated approach that's very
28:09
opinionated approach that's very different from what Google and Amazon
28:11
different from what Google and Amazon
28:11
different from what Google and Amazon did in order to create a big speak
28:14
did in order to create a big speak
28:14
did in order to create a big speak capsule you must use Bigsby studio which
28:17
capsule you must use Bigsby studio which
28:17
capsule you must use Bigsby studio which is an attack which is a completely
28:18
is an attack which is a completely
28:18
is an attack which is a completely integrated development environment the
28:22
integrated development environment the
28:22
integrated development environment the when you create a capsule what they do
28:25
when you create a capsule what they do
28:25
when you create a capsule what they do is they follow the they follow an MVC
28:28
is they follow the they follow an MVC
28:28
is they follow the they follow an MVC pattern so they just renamed it slightly
28:31
pattern so they just renamed it slightly
28:31
pattern so they just renamed it slightly so what you do is you create what they
28:35
so what you do is you create what they
28:35
so what you do is you create what they call concepts that's effectively your
28:38
call concepts that's effectively your
28:38
call concepts that's effectively your model views are called the our views and
28:40
model views are called the our views and
28:40
model views are called the our views and actions are and actions are essentially
28:44
actions are and actions are essentially
28:44
actions are and actions are essentially your controllers one of the one of the
28:48
your controllers one of the one of the
28:48
your controllers one of the one of the things about Samsung is because they
28:50
things about Samsung is because they
28:50
things about Samsung is because they started with the phone instead of echo
28:54
started with the phone instead of echo
28:54
started with the phone instead of echo or Google assistant where there isn't
28:56
or Google assistant where there isn't
28:56
or Google assistant where there isn't necessarily a screen device they go off
28:58
necessarily a screen device they go off
28:58
necessarily a screen device they go off the assumption that you've got a screen
29:00
the assumption that you've got a screen
29:00
the assumption that you've got a screen in front of you somewhere so they really
29:03
in front of you somewhere so they really
29:03
in front of you somewhere so they really work with you to want you to create
29:07
work with you to want you to create
29:07
work with you to want you to create multimedia experiences that are also
29:09
multimedia experiences that are also
29:09
multimedia experiences that are also voiced during
29:11
voiced during
29:11
voiced during which is you know which again is
29:13
which is you know which again is
29:13
which is you know which again is somewhat different but it does make for
29:16
somewhat different but it does make for
29:16
somewhat different but it does make for some pretty cool stuff because you you
29:18
some pretty cool stuff because you you
29:18
some pretty cool stuff because you you end up having to put every you know
29:20
end up having to put every you know
29:20
end up having to put every you know spend a lot of time putting that
29:21
spend a lot of time putting that
29:21
spend a lot of time putting that together in the studio now when you
29:25
together in the studio now when you
29:25
together in the studio now when you define actions they can be defined as
29:28
define actions they can be defined as
29:28
define actions they can be defined as either local or remote if they're local
29:30
either local or remote if they're local
29:30
either local or remote if they're local they're essentially just little chunks
29:34
they're essentially just little chunks
29:34
they're essentially just little chunks of JavaScript that gets shipped with
29:35
of JavaScript that gets shipped with
29:35
of JavaScript that gets shipped with your capsule and end up being hosted by
29:37
your capsule and end up being hosted by
29:37
your capsule and end up being hosted by Samsung if they're remote then each of
29:42
Samsung if they're remote then each of
29:42
Samsung if they're remote then each of those actions is an individual REST API
29:45
those actions is an individual REST API
29:45
those actions is an individual REST API call with Google and Google assistant
29:50
call with Google and Google assistant
29:50
call with Google and Google assistant and Alexa you have a single endpoint
29:52
and Alexa you have a single endpoint
29:53
and Alexa you have a single endpoint that gets a message that you work with
29:55
that gets a message that you work with
29:55
that gets a message that you work with in the case of a big speed capsule each
29:59
in the case of a big speed capsule each
29:59
in the case of a big speed capsule each one of your actions has to have its own
30:03
one of your actions has to have its own
30:03
one of your actions has to have its own endpoint so you know that's just
30:06
endpoint so you know that's just
30:06
endpoint so you know that's just something else to consider and one of
30:09
something else to consider and one of
30:09
something else to consider and one of the other things that comes with this
30:11
the other things that comes with this
30:11
the other things that comes with this model is that when you publish the
30:12
model is that when you publish the
30:12
model is that when you publish the capsule everything gets pushed and
30:14
capsule everything gets pushed and
30:14
capsule everything gets pushed and hosted up in Samsung's cloud but because
30:18
hosted up in Samsung's cloud but because
30:18
hosted up in Samsung's cloud but because everything is so tightly bound any sort
30:22
everything is so tightly bound any sort
30:22
everything is so tightly bound any sort of a change even if it's a minor one
30:24
of a change even if it's a minor one
30:24
of a change even if it's a minor one like just updating an image and things
30:26
like just updating an image and things
30:26
like just updating an image and things like or changing a response you pretty
30:30
like or changing a response you pretty
30:30
like or changing a response you pretty much end up having to republish the
30:31
much end up having to republish the
30:31
much end up having to republish the entire capsule with Google and Alexa as
30:37
entire capsule with Google and Alexa as
30:37
entire capsule with Google and Alexa as long as you're not making any
30:39
long as you're not making any
30:39
long as you're not making any significant changes to the natural
30:41
significant changes to the natural
30:41
significant changes to the natural language model you can actually make
30:44
language model you can actually make
30:44
language model you can actually make make changes to responses and things
30:46
make changes to responses and things
30:46
make changes to responses and things like that without without having to do
30:49
like that without without having to do
30:49
like that without without having to do it do an entirely new publish so on that
30:54
it do an entirely new publish so on that
30:54
it do an entirely new publish so on that note I'm gonna switch over to talking
30:57
note I'm gonna switch over to talking
30:57
note I'm gonna switch over to talking about future I did see somebody asked
31:00
about future I did see somebody asked
31:00
about future I did see somebody asked about voice SEO so you know that so I
31:04
about voice SEO so you know that so I
31:04
about voice SEO so you know that so I what I'd like to do is start by talking
31:07
what I'd like to do is start by talking
31:07
what I'd like to do is start by talking about search and discoverability the
31:10
about search and discoverability the
31:10
about search and discoverability the reality is right now on these devices
31:13
reality is right now on these devices
31:13
reality is right now on these devices there's just really no good story one of
31:17
there's just really no good story one of
31:17
there's just really no good story one of the one of the challenges and the that
31:20
the one of the challenges and the that
31:20
the one of the challenges and the that has happened with the with the way that
31:22
has happened with the with the way that
31:22
has happened with the with the way that this is sort of sprung
31:24
this is sort of sprung
31:24
this is sort of sprung is that this is kind of grown
31:26
is that this is kind of grown
31:26
is that this is kind of grown organically you had you know Apple
31:29
organically you had you know Apple
31:29
organically you had you know Apple started with Siri you have Alexa with
31:32
started with Siri you have Alexa with
31:32
started with Siri you have Alexa with Amazon you have Google assistant you got
31:35
Amazon you have Google assistant you got
31:35
Amazon you have Google assistant you got say all of these are walled gardens
31:37
say all of these are walled gardens
31:37
say all of these are walled gardens everybody's doing things differently and
31:42
everybody's doing things differently and
31:42
everybody's doing things differently and you know it so so you know what if you
31:45
you know it so so you know what if you
31:45
you know it so so you know what if you want something to work on all three of
31:47
want something to work on all three of
31:47
want something to work on all three of these different devices you end up
31:48
these different devices you end up
31:48
these different devices you end up having to having to do different
31:50
having to having to do different
31:50
having to having to do different implementations it's you know and one of
31:54
implementations it's you know and one of
31:54
implementations it's you know and one of the things with search and
31:54
the things with search and
31:54
the things with search and discoverability is they all do it
31:56
discoverability is they all do it
31:57
discoverability is they all do it differently so for example with a Lex
32:01
differently so for example with a Lex
32:01
differently so for example with a Lex and Google they actually had in Canada
32:04
and Google they actually had in Canada
32:04
and Google they actually had in Canada earlier about a year ago they were
32:06
earlier about a year ago they were
32:06
earlier about a year ago they were really even with the Amazon I think was
32:08
really even with the Amazon I think was
32:08
really even with the Amazon I think was about two years ago they had the notion
32:11
about two years ago they had the notion
32:11
about two years ago they had the notion of something that they called a can
32:13
of something that they called a can
32:13
of something that they called a can fulfill intent request which was
32:15
fulfill intent request which was
32:15
fulfill intent request which was something where you could say I support
32:18
something where you could say I support
32:18
something where you could say I support this interface and if Alexa thought that
32:22
this interface and if Alexa thought that
32:22
this interface and if Alexa thought that a question might apply to your skill
32:26
a question might apply to your skill
32:26
a question might apply to your skill then it would call your call you and say
32:29
then it would call your call you and say
32:29
then it would call your call you and say hey can you answer this and that you
32:33
hey can you answer this and that you
32:33
hey can you answer this and that you know that was an idea it's never left
32:35
know that was an idea it's never left
32:35
know that was an idea it's never left beta it's still available but I've never
32:37
beta it's still available but I've never
32:37
beta it's still available but I've never actually seen it work and then with
32:42
actually seen it work and then with
32:42
actually seen it work and then with Google assistant they had the notion of
32:44
Google assistant they had the notion of
32:45
Google assistant they had the notion of indirect invocation which is similar to
32:47
indirect invocation which is similar to
32:47
indirect invocation which is similar to the can fulfill intent request where you
32:49
the can fulfill intent request where you
32:49
the can fulfill intent request where you could specify here's a bunch of phrases
32:52
could specify here's a bunch of phrases
32:52
could specify here's a bunch of phrases that if if you get asked I know what to
32:56
that if if you get asked I know what to
32:56
that if if you get asked I know what to do with that seem to be a direction they
32:59
do with that seem to be a direction they
32:59
do with that seem to be a direction they were going down about a year ago but
33:02
were going down about a year ago but
33:02
were going down about a year ago but then you know they kind of both the both
33:04
then you know they kind of both the both
33:04
then you know they kind of both the both the platforms seem to have stepped away
33:06
the platforms seem to have stepped away
33:06
the platforms seem to have stepped away from it as regards the voice SEO
33:09
from it as regards the voice SEO
33:09
from it as regards the voice SEO question what Google is done is and you
33:14
question what Google is done is and you
33:14
question what Google is done is and you know there's plenty of resources on the
33:15
know there's plenty of resources on the
33:15
know there's plenty of resources on the internet now I mean if you look up
33:17
internet now I mean if you look up
33:17
internet now I mean if you look up Google assistant and SEO they have gone
33:20
Google assistant and SEO they have gone
33:20
Google assistant and SEO they have gone down this path of looking for structured
33:23
down this path of looking for structured
33:23
down this path of looking for structured content in websites that can be used to
33:26
content in websites that can be used to
33:26
content in websites that can be used to inform search you know Google search you
33:31
inform search you know Google search you
33:31
inform search you know Google search you know what what what to do if they're
33:33
know what what what to do if they're
33:33
know what what what to do if they're being at if it's being accessed through
33:36
being at if it's being accessed through
33:36
being at if it's being accessed through a voice assistant and
33:38
a voice assistant and
33:38
a voice assistant and that's you know that seems like a
33:39
that's you know that seems like a
33:39
that's you know that seems like a reasonable strategy the downside to that
33:43
reasonable strategy the downside to that
33:43
reasonable strategy the downside to that is that it's not like a Google action
33:46
is that it's not like a Google action
33:46
is that it's not like a Google action it's not really interactive you know so
33:49
it's not really interactive you know so
33:49
it's not really interactive you know so you get something there so I you know I
33:51
you get something there so I you know I
33:51
you get something there so I you know I don't know what the story is yet I don't
33:53
don't know what the story is yet I don't
33:53
don't know what the story is yet I don't think Google knows what those furry is
33:54
think Google knows what those furry is
33:54
think Google knows what those furry is yet how actions are going to fit into
33:56
yet how actions are going to fit into
33:56
yet how actions are going to fit into this and I think you see this you'll see
33:58
this and I think you see this you'll see
33:58
this and I think you see this you'll see the same thing with Amazon I haven't
34:01
the same thing with Amazon I haven't
34:01
the same thing with Amazon I haven't heard much about what they do as far as
34:03
heard much about what they do as far as
34:03
heard much about what they do as far as as far as structured content I do know
34:07
as far as structured content I do know
34:07
as far as structured content I do know that they've been going down this path
34:08
that they've been going down this path
34:08
that they've been going down this path of Alexa answers where they're
34:11
of Alexa answers where they're
34:11
of Alexa answers where they're crowdsourcing content a lot of times
34:14
crowdsourcing content a lot of times
34:14
crowdsourcing content a lot of times when they search that they end up
34:16
when they search that they end up
34:16
when they search that they end up hitting Wikipedia to try to get answers
34:19
hitting Wikipedia to try to get answers
34:19
hitting Wikipedia to try to get answers but that's suboptimal so they've been
34:23
but that's suboptimal so they've been
34:23
but that's suboptimal so they've been going through a crowd sourcing model on
34:25
going through a crowd sourcing model on
34:25
going through a crowd sourcing model on that but that's you know that's kind of
34:28
that but that's you know that's kind of
34:28
that but that's you know that's kind of where this is this is emerging
34:30
where this is this is emerging
34:30
where this is this is emerging technology so a lot of this is literally
34:32
technology so a lot of this is literally
34:32
technology so a lot of this is literally being figured out is still being figured
34:35
being figured out is still being figured
34:35
being figured out is still being figured out all right
34:37
out all right
34:37
out all right next up there's there's the note notion
34:39
next up there's there's the note notion
34:39
next up there's there's the note notion of seamless integrations I mean the
34:41
of seamless integrations I mean the
34:41
of seamless integrations I mean the current reliance on indication names for
34:44
current reliance on indication names for
34:44
current reliance on indication names for a voice application is kind of clunky
34:47
a voice application is kind of clunky
34:47
a voice application is kind of clunky and one of the things we've seen and we
34:50
and one of the things we've seen and we
34:50
and one of the things we've seen and we can see this actually from interactions
34:52
can see this actually from interactions
34:52
can see this actually from interactions that you know John talked about on how
34:55
that you know John talked about on how
34:55
that you know John talked about on how Google we actually see what people have
34:57
Google we actually see what people have
34:57
Google we actually see what people have said so we've seen how people have
35:01
said so we've seen how people have
35:01
said so we've seen how people have interacted with some of the applications
35:04
interacted with some of the applications
35:04
interacted with some of the applications we've done and it's become it becomes
35:07
we've done and it's become it becomes
35:07
we've done and it's become it becomes really clear that people don't
35:09
really clear that people don't
35:09
really clear that people don't necessarily know that they're in a voice
35:13
necessarily know that they're in a voice
35:13
necessarily know that they're in a voice application that you know you've
35:15
application that you know you've
35:15
application that you know you've actually transferred control from the
35:17
actually transferred control from the
35:17
actually transferred control from the assistant to the application so you know
35:21
assistant to the application so you know
35:21
assistant to the application so you know some people kind of some applications
35:23
some people kind of some applications
35:23
some people kind of some applications will work around that by check you know
35:25
will work around that by check you know
35:25
will work around that by check you know this is one of the reasons people change
35:26
this is one of the reasons people change
35:26
this is one of the reasons people change voices or use mp3 files because it does
35:30
voices or use mp3 files because it does
35:30
voices or use mp3 files because it does make it more clear you're within the
35:32
make it more clear you're within the
35:32
make it more clear you're within the world of an application but a lot of
35:34
world of an application but a lot of
35:34
world of an application but a lot of users don't get that I think as we start
35:37
users don't get that I think as we start
35:37
users don't get that I think as we start going forwards the the delineation
35:41
going forwards the the delineation
35:41
going forwards the the delineation between the assistant and an application
35:44
between the assistant and an application
35:44
between the assistant and an application will eventually start to start to you
35:49
will eventually start to start to you
35:49
will eventually start to start to you know blur
35:51
know blur
35:51
know blur we're applications maybe more hints to
35:53
we're applications maybe more hints to
35:53
we're applications maybe more hints to the assistant but you know that's
35:55
the assistant but you know that's
35:55
the assistant but you know that's something that's something I think that
35:56
something that's something I think that
35:56
something that's something I think that is eventually going to play out again
35:58
is eventually going to play out again
35:58
is eventually going to play out again we're we're we're dealing with emerging
36:01
we're we're we're dealing with emerging
36:01
we're we're we're dealing with emerging technologies so some of this is is still
36:04
technologies so some of this is is still
36:04
technologies so some of this is is still being defined and John what I'll turn it
36:07
being defined and John what I'll turn it
36:07
being defined and John what I'll turn it over to you now for the last - sure so
36:10
over to you now for the last - sure so
36:10
over to you now for the last - sure so before I do dive into those I do see a
36:13
before I do dive into those I do see a
36:13
before I do dive into those I do see a question on it when can we have an
36:16
question on it when can we have an
36:16
question on it when can we have an assistant vest enough to reply to our
36:17
assistant vest enough to reply to our
36:17
assistant vest enough to reply to our commands instantly some of the
36:19
commands instantly some of the
36:19
commands instantly some of the technology push technology depends on
36:21
technology push technology depends on
36:21
technology push technology depends on servers is not fast enough so that
36:24
servers is not fast enough so that
36:24
servers is not fast enough so that speaks to a an OEM embedded model where
36:30
speaks to a an OEM embedded model where
36:30
speaks to a an OEM embedded model where the assistant is embedded in a device
36:33
the assistant is embedded in a device
36:33
the assistant is embedded in a device that is local to the user we have seen
36:39
that is local to the user we have seen
36:39
that is local to the user we have seen that in you know we've gone to a number
36:41
that in you know we've gone to a number
36:41
that in you know we've gone to a number of conferences and voice technology and
36:44
of conferences and voice technology and
36:44
of conferences and voice technology and see some technology that's emerging in
36:47
see some technology that's emerging in
36:47
see some technology that's emerging in in automobiles right it's trying to
36:52
in automobiles right it's trying to
36:52
in automobiles right it's trying to become a bit more common place to find
36:55
become a bit more common place to find
36:55
become a bit more common place to find integrated voice experiences in you know
36:59
integrated voice experiences in you know
36:59
integrated voice experiences in you know Toyota I believe has their current year
37:02
Toyota I believe has their current year
37:02
Toyota I believe has their current year models included voice assistance so you
37:05
models included voice assistance so you
37:05
models included voice assistance so you would be able to change your your radio
37:11
would be able to change your your radio
37:11
would be able to change your your radio stations for example by voice Mercedes
37:14
stations for example by voice Mercedes
37:14
stations for example by voice Mercedes had one of the conference's we went to
37:17
had one of the conference's we went to
37:17
had one of the conference's we went to had a Mercedes on display that would let
37:20
had a Mercedes on display that would let
37:20
had a Mercedes on display that would let us change you know lights inside the car
37:23
us change you know lights inside the car
37:23
us change you know lights inside the car to different colors it would let us
37:25
to different colors it would let us
37:25
to different colors it would let us change the EC settings
37:29
change the EC settings
37:29
change the EC settings interestingly enough and this is a just
37:32
interestingly enough and this is a just
37:32
interestingly enough and this is a just a footnote but it wouldn't let us raise
37:35
a footnote but it wouldn't let us raise
37:35
a footnote but it wouldn't let us raise or lower the windows interestingly
37:37
or lower the windows interestingly
37:37
or lower the windows interestingly enough because you can't communicate how
37:40
enough because you can't communicate how
37:40
enough because you can't communicate how far you want the window to go up or down
37:43
far you want the window to go up or down
37:43
far you want the window to go up or down easily in a voice command
37:45
easily in a voice command
37:45
easily in a voice command it doesn't sound natural to say you know
37:47
it doesn't sound natural to say you know
37:47
it doesn't sound natural to say you know lower window ten percent or eighty
37:50
lower window ten percent or eighty
37:50
lower window ten percent or eighty percent it's the command would be you
37:52
percent it's the command would be you
37:52
percent it's the command would be you know lower window and the auto
37:54
know lower window and the auto
37:54
know lower window and the auto manufacturers are still working on the
37:56
manufacturers are still working on the
37:56
manufacturers are still working on the nuances of how to interpret some of
37:59
nuances of how to interpret some of
37:59
nuances of how to interpret some of those commands
38:02
now in some cases the servers can be a
38:08
now in some cases the servers can be a
38:08
now in some cases the servers can be a little slow in starting up so if your
38:10
little slow in starting up so if your
38:10
little slow in starting up so if your this goes back to the earlier design
38:13
this goes back to the earlier design
38:13
this goes back to the earlier design questions and where design
38:14
questions and where design
38:14
questions and where design considerations around standing up a
38:16
considerations around standing up a
38:16
considerations around standing up a skill around a lambda function sometimes
38:21
skill around a lambda function sometimes
38:21
skill around a lambda function sometimes you'll find a cold start with your
38:24
you'll find a cold start with your
38:24
you'll find a cold start with your lambda function and once it starts up
38:27
lambda function and once it starts up
38:27
lambda function and once it starts up it's much more responsive and the
38:30
it's much more responsive and the
38:30
it's much more responsive and the response rates from warm lambda or a
38:34
response rates from warm lambda or a
38:34
response rates from warm lambda or a worm you know as your function or nearly
38:38
worm you know as your function or nearly
38:38
worm you know as your function or nearly instantaneous but the first hit to a
38:41
instantaneous but the first hit to a
38:41
instantaneous but the first hit to a cold function could incur a couple
38:45
cold function could incur a couple
38:45
cold function could incur a couple seconds of start-up time and there you
38:47
seconds of start-up time and there you
38:47
seconds of start-up time and there you want to minimize the number of resources
38:49
want to minimize the number of resources
38:49
want to minimize the number of resources that your service function depends on
38:52
that your service function depends on
38:52
that your service function depends on when it bootstraps so rather than say
38:54
when it bootstraps so rather than say
38:54
when it bootstraps so rather than say going to a configuration story you may
38:57
going to a configuration story you may
38:57
going to a configuration story you may want to embed the configuration into
38:59
want to embed the configuration into
38:59
want to embed the configuration into your deployed lambda oh and reduce the
39:03
your deployed lambda oh and reduce the
39:03
your deployed lambda oh and reduce the the number of resources that tends on in
39:07
the number of resources that tends on in
39:07
the number of resources that tends on in in one startup so with respect to touch
39:13
in one startup so with respect to touch
39:13
in one startup so with respect to touch this interfaces moving back to the slide
39:17
this interfaces moving back to the slide
39:17
this interfaces moving back to the slide of course the current pandemic is on
39:20
of course the current pandemic is on
39:20
of course the current pandemic is on everybody's mind and technologists are
39:24
everybody's mind and technologists are
39:24
everybody's mind and technologists are involved in voice technology are are
39:30
involved in voice technology are are
39:30
involved in voice technology are are considering and thinking about how to
39:33
considering and thinking about how to
39:33
considering and thinking about how to leverage voice technology to assist in
39:36
leverage voice technology to assist in
39:36
leverage voice technology to assist in reducing the the spread of koban 19 and
39:41
reducing the the spread of koban 19 and
39:41
reducing the the spread of koban 19 and this moves beyond the smart speaker
39:45
this moves beyond the smart speaker
39:45
this moves beyond the smart speaker experience of course you can get kovat
39:47
experience of course you can get kovat
39:47
experience of course you can get kovat 19 updates on alexa on google the
39:51
19 updates on alexa on google the
39:51
19 updates on alexa on google the general rate of people interacting that
39:55
general rate of people interacting that
39:55
general rate of people interacting that support speakers has gone up more people
39:57
support speakers has gone up more people
39:57
support speakers has gone up more people are home more people are trying them and
39:59
are home more people are trying them and
39:59
are home more people are trying them and more people are engaging with them there
40:02
more people are engaging with them there
40:02
more people are engaging with them there there we've certainly seen an uptick in
40:04
there we've certainly seen an uptick in
40:04
there we've certainly seen an uptick in this voice applications we've released
40:07
this voice applications we've released
40:07
this voice applications we've released and and that's not a and we're certainly
40:11
and and that's not a and we're certainly
40:11
and and that's not a and we're certainly not a unique that's not unique situation
40:16
not a unique that's not unique situation
40:16
not a unique that's not unique situation but I think we're going to see voice
40:21
but I think we're going to see voice
40:21
but I think we're going to see voice experiences move into other unexpected
40:25
experiences move into other unexpected
40:25
experiences move into other unexpected at least in the u.s. unexpected devices
40:29
at least in the u.s. unexpected devices
40:29
at least in the u.s. unexpected devices for example in Japan you can get into an
40:32
for example in Japan you can get into an
40:32
for example in Japan you can get into an elevator and you don't have to touch a
40:33
elevator and you don't have to touch a
40:33
elevator and you don't have to touch a button you can ask you can tell it what
40:35
button you can ask you can tell it what
40:35
button you can ask you can tell it what floor to go to and I anticipate that
40:38
floor to go to and I anticipate that
40:38
floor to go to and I anticipate that that technology will be coming to the US
40:41
that technology will be coming to the US
40:41
that technology will be coming to the US and other countries where you want to
40:47
and other countries where you want to
40:47
and other countries where you want to take some care is when you're
40:50
take some care is when you're
40:50
take some care is when you're considering what kiosks and public voice
40:55
considering what kiosks and public voice
40:55
considering what kiosks and public voice experiences to enable of course you want
40:58
experiences to enable of course you want
40:58
experiences to enable of course you want to conserve security so voice enabling
41:01
to conserve security so voice enabling
41:01
to conserve security so voice enabling an ATM for example may only be available
41:05
an ATM for example may only be available
41:05
an ATM for example may only be available if that ATM is located in a booth and
41:08
if that ATM is located in a booth and
41:08
if that ATM is located in a booth and there's only one person expected to be
41:10
there's only one person expected to be
41:10
there's only one person expected to be in that booth so you don't over here say
41:12
in that booth so you don't over here say
41:12
in that booth so you don't over here say a pin number or any sensitive financial
41:15
a pin number or any sensitive financial
41:15
a pin number or any sensitive financial data so there's of course there's a
41:25
data so there's of course there's a
41:25
data so there's of course there's a concern about security in general with
41:27
concern about security in general with
41:27
concern about security in general with voice technology you know a common
41:29
voice technology you know a common
41:29
voice technology you know a common objection that consumers have for not
41:32
objection that consumers have for not
41:32
objection that consumers have for not adopting smart speakers is that they
41:34
adopting smart speakers is that they
41:34
adopting smart speakers is that they don't want Alexa or Amazon or Google to
41:38
don't want Alexa or Amazon or Google to
41:38
don't want Alexa or Amazon or Google to hold on to their voice samples and you
41:42
hold on to their voice samples and you
41:42
hold on to their voice samples and you know the companies have been coming
41:44
know the companies have been coming
41:44
know the companies have been coming around and giving users privacy settings
41:47
around and giving users privacy settings
41:47
around and giving users privacy settings that enable you to not store you were
41:50
that enable you to not store you were
41:50
that enable you to not store you were your voice history but there's been an
41:54
your voice history but there's been an
41:54
your voice history but there's been an organization that's emerging called the
41:57
organization that's emerging called the
41:57
organization that's emerging called the open voice Network that's attempting to
42:00
open voice Network that's attempting to
42:00
open voice Network that's attempting to bring a level of trust to smart speakers
42:04
bring a level of trust to smart speakers
42:04
bring a level of trust to smart speakers and voice you UX experiences so they're
42:10
and voice you UX experiences so they're
42:10
and voice you UX experiences so they're focusing on security they're trying
42:13
focusing on security they're trying
42:13
focusing on security they're trying their the best analogy I can think of
42:17
their the best analogy I can think of
42:17
their the best analogy I can think of for the open voice network is they're
42:19
for the open voice network is they're
42:19
for the open voice network is they're trying they're becoming a w3c they're
42:23
trying they're becoming a w3c they're
42:23
trying they're becoming a w3c they're there to let me come up with a analogy
42:28
there
42:29
there
42:29
there what HTTP and the web is to w3c is what
42:36
what HTTP and the web is to w3c is what
42:36
what HTTP and the web is to w3c is what open voice Network is targeting were
42:39
open voice Network is targeting were
42:39
open voice Network is targeting were voice applications and smart speakers so
42:43
voice applications and smart speakers so
42:43
voice applications and smart speakers so that includes standardization around
42:46
that includes standardization around
42:46
that includes standardization around security as well as creating a one of
42:51
security as well as creating a one of
42:51
security as well as creating a one of the early mandates for the open voice
42:53
the early mandates for the open voice
42:53
the early mandates for the open voice network is to create effectively a voice
42:57
network is to create effectively a voice
42:57
network is to create effectively a voice domain name service for invocation names
43:00
domain name service for invocation names
43:00
domain name service for invocation names so that for example if you go to if you
43:03
so that for example if you go to if you
43:03
so that for example if you go to if you expect to talk to McDonald's on on Alexa
43:07
expect to talk to McDonald's on on Alexa
43:07
expect to talk to McDonald's on on Alexa you would use the same the same command
43:12
you would use the same the same command
43:12
you would use the same the same command open a McDonald's Atwood and that would
43:14
open a McDonald's Atwood and that would
43:14
open a McDonald's Atwood and that would be the same command on Google and the
43:15
be the same command on Google and the
43:15
be the same command on Google and the same command on Bigsby so the state of
43:20
same command on Bigsby so the state of
43:20
same command on Bigsby so the state of woosh technology today is a little bit
43:23
woosh technology today is a little bit
43:23
woosh technology today is a little bit like it was in you know in the 90s with
43:26
like it was in you know in the 90s with
43:26
like it was in you know in the 90s with websites where you didn't yet have
43:29
websites where you didn't yet have
43:29
websites where you didn't yet have search engines if you wanted to look for
43:30
search engines if you wanted to look for
43:31
search engines if you wanted to look for a website you would go to some kind of a
43:32
a website you would go to some kind of a
43:32
a website you would go to some kind of a website aggregator like yahoo this is
43:36
website aggregator like yahoo this is
43:36
website aggregator like yahoo this is before google came along so the open
43:39
before google came along so the open
43:39
before google came along so the open voice network is trying to bring some
43:42
voice network is trying to bring some
43:42
voice network is trying to bring some consistency to the what's now the wild
43:45
consistency to the what's now the wild
43:45
consistency to the what's now the wild west of voice applications okay but
43:54
west of voice applications okay but
43:54
west of voice applications okay but that's that's perfect I think you guys
43:56
that's that's perfect I think you guys
43:56
that's that's perfect I think you guys covered a lot of content in the first
43:59
covered a lot of content in the first
43:59
covered a lot of content in the first half maybe right there you talked about
44:02
half maybe right there you talked about
44:02
half maybe right there you talked about Alexa the difference between all these
44:05
Alexa the difference between all these
44:05
Alexa the difference between all these three big specv and all also from the
44:08
three big specv and all also from the
44:08
three big specv and all also from the developers perfects they've the
44:10
developers perfects they've the
44:10
developers perfects they've the documentation z' that are available for
44:12
documentation z' that are available for
44:12
documentation z' that are available for all these platforms a couple of
44:15
all these platforms a couple of
44:15
all these platforms a couple of questions that you did indeed pick some
44:17
questions that you did indeed pick some
44:17
questions that you did indeed pick some of the questions from the comments a
44:19
of the questions from the comments a
44:19
of the questions from the comments a couple of questions that heads here is a
44:22
couple of questions that heads here is a
44:22
couple of questions that heads here is a one question that says what is the
44:26
one question that says what is the
44:26
one question that says what is the potential of jobs in voice tech as
44:28
potential of jobs in voice tech as
44:28
potential of jobs in voice tech as compared to the normal tech like a web
44:30
compared to the normal tech like a web
44:30
compared to the normal tech like a web tab because you know we have seen very
44:32
tab because you know we have seen very
44:32
tab because you know we have seen very this fancy JavaScript framework coming
44:35
this fancy JavaScript framework coming
44:35
this fancy JavaScript framework coming and web development has taken a really
44:37
and web development has taken a really
44:37
and web development has taken a really new phase in from 2017 and 18 and since
44:42
new phase in from 2017 and 18 and since
44:42
new phase in from 2017 and 18 and since his voice technology is very
44:43
his voice technology is very
44:43
his voice technology is very you now how should one approach and what
44:48
you now how should one approach and what
44:48
you now how should one approach and what should be the progress in this journey
44:51
should be the progress in this journey
44:51
should be the progress in this journey yeah I can I can start that once on
44:54
yeah I can I can start that once on
44:54
yeah I can I can start that once on chime in because I do have a couple of
44:57
chime in because I do have a couple of
44:57
chime in because I do have a couple of thoughts on that so what we've seen in
45:00
thoughts on that so what we've seen in
45:00
thoughts on that so what we've seen in Bush technology from the conference's
45:02
Bush technology from the conference's
45:02
Bush technology from the conference's that we've gone to is and I Baloo did to
45:08
that we've gone to is and I Baloo did to
45:08
that we've gone to is and I Baloo did to it earlier voice user experience design
45:10
it earlier voice user experience design
45:10
it earlier voice user experience design is a discipline that's been emerging out
45:15
is a discipline that's been emerging out
45:15
is a discipline that's been emerging out of the Wild West of voice application
45:19
of the Wild West of voice application
45:19
of the Wild West of voice application development and it's trying to bring a
45:23
development and it's trying to bring a
45:23
development and it's trying to bring a set of standards and practices like just
45:26
set of standards and practices like just
45:26
set of standards and practices like just as you had used UX user experience
45:29
as you had used UX user experience
45:29
as you had used UX user experience design has grown and evolved over the
45:33
design has grown and evolved over the
45:33
design has grown and evolved over the last decade voice UX and voice user
45:36
last decade voice UX and voice user
45:36
last decade voice UX and voice user experience design is starting to emerge
45:39
experience design is starting to emerge
45:39
experience design is starting to emerge as a dedicated discipline and that
45:44
as a dedicated discipline and that
45:44
as a dedicated discipline and that includes designing for accessibility
45:49
includes designing for accessibility
45:49
includes designing for accessibility designing for people that may have may
45:52
designing for people that may have may
45:52
designing for people that may have may be a bit hard of hearing how do you
45:54
be a bit hard of hearing how do you
45:54
be a bit hard of hearing how do you design the experience for that and of
45:58
design the experience for that and of
45:58
design the experience for that and of course there's you know targeted
46:02
course there's you know targeted
46:02
course there's you know targeted specialized development in this space so
46:06
specialized development in this space so
46:06
specialized development in this space so you've got your designers and got your
46:08
you've got your designers and got your
46:08
you've got your designers and got your voice experienced developers saunch do
46:11
voice experienced developers saunch do
46:11
voice experienced developers saunch do you have anything to know the other
46:13
you have anything to know the other
46:13
you have anything to know the other thing I would definitely add this is
46:16
thing I would definitely add this is
46:16
thing I would definitely add this is still emerging so the the the number of
46:20
still emerging so the the the number of
46:20
still emerging so the the the number of people that are in here you know is it's
46:23
people that are in here you know is it's
46:23
people that are in here you know is it's it's still pretty still relatively small
46:26
it's still pretty still relatively small
46:26
it's still pretty still relatively small so when people are looking for when
46:30
so when people are looking for when
46:30
so when people are looking for when people are looking for voice UX and
46:32
people are looking for voice UX and
46:32
people are looking for voice UX and developers that there's not a big pool
46:34
developers that there's not a big pool
46:34
developers that there's not a big pool of people to pick from so you know if
46:37
of people to pick from so you know if
46:37
of people to pick from so you know if you do get good at this and start
46:40
you do get good at this and start
46:40
you do get good at this and start putting yourself out there there is
46:42
putting yourself out there there is
46:42
putting yourself out there there is quite a bit of opportunity and you know
46:43
quite a bit of opportunity and you know
46:43
quite a bit of opportunity and you know if you were at the beginning of the
46:45
if you were at the beginning of the
46:45
if you were at the beginning of the presentation you can see the numbers you
46:47
presentation you can see the numbers you
46:47
presentation you can see the numbers you know they do speak to something you know
46:50
know they do speak to something you know
46:50
know they do speak to something you know somewhere you're gonna see this stuff
46:52
somewhere you're gonna see this stuff
46:52
somewhere you're gonna see this stuff start to take off it's not a question of
46:54
start to take off it's not a question of
46:54
start to take off it's not a question of if it's going to take off it's when it
46:56
if it's going to take off it's when it
46:56
if it's going to take off it's when it does
46:57
does
46:57
does I think an excellent place to start is
47:01
I think an excellent place to start is
47:01
I think an excellent place to start is that one of the things amazon has done
47:04
that one of the things amazon has done
47:04
that one of the things amazon has done is they have an Alexis skill developer
47:07
is they have an Alexis skill developer
47:07
is they have an Alexis skill developer certification so you can become you know
47:10
certification so you can become you know
47:10
certification so you can become you know you know we we actually have both been
47:13
you know we we actually have both been
47:13
you know we we actually have both been through that and actually John has been
47:16
through that and actually John has been
47:16
through that and actually John has been John was about a year ago wasn't helping
47:19
John was about a year ago wasn't helping
47:19
John was about a year ago wasn't helping write some of the cooking questions for
47:21
write some of the cooking questions for
47:21
write some of the cooking questions for the newer tests that's right so if you
47:23
the newer tests that's right so if you
47:23
the newer tests that's right so if you do you take the Amazon lots of skill
47:25
do you take the Amazon lots of skill
47:25
do you take the Amazon lots of skill certification and make it some of my
47:26
certification and make it some of my
47:26
certification and make it some of my questions yeah that really hard that
47:33
questions yeah that really hard that
47:33
questions yeah that really hard that does well answered John and Sanjeev
47:36
does well answered John and Sanjeev
47:36
does well answered John and Sanjeev adding to that you guys both run a
47:38
adding to that you guys both run a
47:38
adding to that you guys both run a company out of this technology right and
47:40
company out of this technology right and
47:41
company out of this technology right and you did mention they are a bunch of
47:42
you did mention they are a bunch of
47:42
you did mention they are a bunch of documentation and there's some studios
47:45
documentation and there's some studios
47:45
documentation and there's some studios when you go and start developing for big
47:47
when you go and start developing for big
47:47
when you go and start developing for big speed one cushion that comes from the
47:50
speed one cushion that comes from the
47:50
speed one cushion that comes from the comments is is there any open source
47:52
comments is is there any open source
47:52
comments is is there any open source library available for this voice
47:54
library available for this voice
47:54
library available for this voice technology you said this field is very
47:56
technology you said this field is very
47:56
technology you said this field is very niche now or not many people are into
47:59
niche now or not many people are into
47:59
niche now or not many people are into this so is there have you come across
48:01
this so is there have you come across
48:01
this so is there have you come across any open source library all this time I
48:04
any open source library all this time I
48:04
any open source library all this time I can I can start that once on Jan yeah
48:07
can I can start that once on Jan yeah
48:07
can I can start that once on Jan yeah you can feel free to add so so there's
48:13
you can feel free to add so so there's
48:13
you can feel free to add so so there's sometimes there's a challenge if she
48:15
sometimes there's a challenge if she
48:15
sometimes there's a challenge if she wants to create a voice experience for a
48:18
wants to create a voice experience for a
48:18
wants to create a voice experience for a platform that doesn't natively have a an
48:21
platform that doesn't natively have a an
48:21
platform that doesn't natively have a an NLP engine so I'll give it a little more
48:25
NLP engine so I'll give it a little more
48:25
NLP engine so I'll give it a little more concrete
48:25
concrete
48:25
concrete so with Alexa and Google right they they
48:29
so with Alexa and Google right they they
48:29
so with Alexa and Google right they they do the natural language processing for
48:31
do the natural language processing for
48:31
do the natural language processing for you you know they do the translation of
48:33
you you know they do the translation of
48:33
you you know they do the translation of the of what the person says to do speech
48:40
the of what the person says to do speech
48:40
the of what the person says to do speech to text transformation and then map that
48:43
to text transformation and then map that
48:43
to text transformation and then map that text to an intent and slot values etc if
48:49
text to an intent and slot values etc if
48:49
text to an intent and slot values etc if you are developing a embedded solution
48:54
you are developing a embedded solution
48:54
you are developing a embedded solution for an embedded device then it's
48:57
for an embedded device then it's
48:57
for an embedded device then it's incumbent upon you to create that NLP
49:00
incumbent upon you to create that NLP
49:00
incumbent upon you to create that NLP model so there is a stanford engine come
49:07
model so there is a stanford engine come
49:07
model so there is a stanford engine come coming out of stanford university that's
49:09
coming out of stanford university that's
49:09
coming out of stanford university that's an open source model for
49:10
an open source model for
49:10
an open source model for doing localized multilingual natural
49:17
doing localized multilingual natural
49:17
doing localized multilingual natural language processing and they have they
49:20
language processing and they have they
49:20
language processing and they have they have a pretty reasonable licensing
49:24
have a pretty reasonable licensing
49:24
have a pretty reasonable licensing licensing cost for that and and Assange
49:32
licensing cost for that and and Assange
49:32
licensing cost for that and and Assange is answering I'm going to try and cool
49:34
is answering I'm going to try and cool
49:34
is answering I'm going to try and cool and see if we can remember what the
49:35
and see if we can remember what the
49:35
and see if we can remember what the exact name of that is and done you can
49:38
exact name of that is and done you can
49:38
exact name of that is and done you can just give that link in the private and
49:40
just give that link in the private and
49:40
just give that link in the private and we would send it to the guys right order
49:42
we would send it to the guys right order
49:42
we would send it to the guys right order right sure
49:44
right sure
49:44
right sure and as we're developing for you know
49:47
and as we're developing for you know
49:47
and as we're developing for you know across platforms for Alexa and Google at
49:51
across platforms for Alexa and Google at
49:51
across platforms for Alexa and Google at the same time because you know they do
49:53
the same time because you know they do
49:53
the same time because you know they do use different you know their JSON
49:57
use different you know their JSON
49:57
use different you know their JSON message which are different yeah the
49:58
message which are different yeah the
49:58
message which are different yeah the lots of request looks different from the
50:00
lots of request looks different from the
50:00
lots of request looks different from the dialog flow requests Samsung Bixby's has
50:04
dialog flow requests Samsung Bixby's has
50:04
dialog flow requests Samsung Bixby's has its own model so that's you know we
50:08
its own model so that's you know we
50:08
its own model so that's you know we we've intended to keep this presentation
50:13
we've intended to keep this presentation
50:13
we've intended to keep this presentation not be a you know a sales pitch for our
50:16
not be a you know a sales pitch for our
50:16
not be a you know a sales pitch for our own engine but at what's known
50:17
own engine but at what's known
50:17
own engine but at what's known technologies we have written our own
50:19
technologies we have written our own
50:19
technologies we have written our own engine to deal with the translation so
50:23
engine to deal with the translation so
50:23
engine to deal with the translation so that it's built on a level of
50:25
that it's built on a level of
50:25
that it's built on a level of abstraction so plugging into another you
50:30
abstraction so plugging into another you
50:30
abstraction so plugging into another you know currently we support Alexa and
50:32
know currently we support Alexa and
50:32
know currently we support Alexa and Google and Bixby is on our roadmap so we
50:35
Google and Bixby is on our roadmap so we
50:35
Google and Bixby is on our roadmap so we consider them just different clients
50:36
consider them just different clients
50:36
consider them just different clients that we can plug into our core engine so
50:40
that we can plug into our core engine so
50:40
that we can plug into our core engine so in answer to that question you know what
50:42
in answer to that question you know what
50:42
in answer to that question you know what libraries do we use for across multiple
50:44
libraries do we use for across multiple
50:44
libraries do we use for across multiple platforms well we brought around and
50:49
platforms well we brought around and
50:49
platforms well we brought around and just to add on to what John was talking
50:51
just to add on to what John was talking
50:51
just to add on to what John was talking about there there is so I know with this
50:55
about there there is so I know with this
50:55
about there there is so I know with this game i and i i'm taking some conjecture
50:58
game i and i i'm taking some conjecture
50:58
game i and i i'm taking some conjecture here the stanford engine there there's
51:01
here the stanford engine there there's
51:01
here the stanford engine there there's an open assistant called almond which i
51:04
an open assistant called almond which i
51:04
an open assistant called almond which i believe is based off of the stanford
51:07
believe is based off of the stanford
51:07
believe is based off of the stanford engine and that's again this is all a
51:10
engine and that's again this is all a
51:10
engine and that's again this is all a lot of this is all open source and there
51:14
lot of this is all open source and there
51:14
lot of this is all open source and there was also at one point and i think it's
51:16
was also at one point and i think it's
51:16
was also at one point and i think it's still out there was a thing called
51:17
still out there was a thing called
51:17
still out there was a thing called Mycroft my CR oft and that is also
51:21
Mycroft my CR oft and that is also
51:21
Mycroft my CR oft and that is also that's fully open source
51:23
that's fully open source
51:24
that's fully open source and that but again that's its own engine
51:26
and that but again that's its own engine
51:26
and that but again that's its own engine John didn't you write at one point I
51:29
John didn't you write at one point I
51:29
John didn't you write at one point I think isn't there a repository up on
51:31
think isn't there a repository up on
51:31
think isn't there a repository up on github for John disappeared yes a judge
51:36
github for John disappeared yes a judge
51:36
github for John disappeared yes a judge just I think he's a he's having some up
51:38
just I think he's a he's having some up
51:38
just I think he's a he's having some up maybe an interest yeah in China's yeah
51:43
maybe an interest yeah in China's yeah
51:43
maybe an interest yeah in China's yeah I'm pretty sure because John has
51:46
I'm pretty sure because John has
51:46
I'm pretty sure because John has actually written articles on doing on
51:50
actually written articles on doing on
51:50
actually written articles on doing on you know real basic articles on how to
51:53
you know real basic articles on how to
51:53
you know real basic articles on how to do using.net core how to create a you
51:58
do using.net core how to create a you
51:58
do using.net core how to create a you know how to create an Alexis skill and
52:01
know how to create an Alexis skill and
52:01
know how to create an Alexis skill and how to create a Google action and I
52:04
how to create a Google action and I
52:04
how to create a Google action and I think the the source for that is
52:06
think the the source for that is
52:06
think the the source for that is actually up on github yeah in fact I do
52:11
actually up on github yeah in fact I do
52:11
actually up on github yeah in fact I do have an article that does show you using
52:13
have an article that does show you using
52:13
have an article that does show you using C sharp and and a sure functions on how
52:17
C sharp and and a sure functions on how
52:17
C sharp and and a sure functions on how to support an Alexa skill on Google
52:19
to support an Alexa skill on Google
52:19
to support an Alexa skill on Google action and that that code in those
52:21
action and that that code in those
52:21
action and that that code in those articles is available in our we have a
52:24
articles is available in our we have a
52:24
articles is available in our we have a public whetstone github repository which
52:26
public whetstone github repository which
52:26
public whetstone github repository which is where you can see those code samples
52:30
is where you can see those code samples
52:30
is where you can see those code samples and also link those articles as Sandra's
52:34
and also link those articles as Sandra's
52:34
and also link those articles as Sandra's answering other questions okay perfect
52:38
answering other questions okay perfect
52:38
answering other questions okay perfect so moments back Sanjeev John ventured
52:41
so moments back Sanjeev John ventured
52:41
so moments back Sanjeev John ventured about the slots right would you like to
52:44
about the slots right would you like to
52:44
about the slots right would you like to go ahead and talk what are slots in
52:45
go ahead and talk what are slots in
52:45
go ahead and talk what are slots in Alexa and how you can use them okay so
52:48
Alexa and how you can use them okay so
52:48
Alexa and how you can use them okay so so here's our you know I'll give an
52:51
so here's our you know I'll give an
52:51
so here's our you know I'll give an example that action that actually the
52:54
example that action that actually the
52:54
example that action that actually the GUP of the Alexus skill I demonstrated
52:57
GUP of the Alexus skill I demonstrated
52:57
GUP of the Alexus skill I demonstrated that is actually making some fairly
53:00
that is actually making some fairly
53:00
that is actually making some fairly heavy use of slots and that so the way I
53:04
heavy use of slots and that so the way I
53:04
heavy use of slots and that so the way I think of slots is they when you create
53:08
think of slots is they when you create
53:08
think of slots is they when you create an utterance you can create a
53:09
an utterance you can create a
53:09
an utterance you can create a placeholder which is effectively a
53:11
placeholder which is effectively a
53:11
placeholder which is effectively a variable that you can say is a slot now
53:13
variable that you can say is a slot now
53:13
variable that you can say is a slot now when you define that slot what you're
53:16
when you define that slot what you're
53:16
when you define that slot what you're really making is an enumeration and
53:19
really making is an enumeration and
53:19
really making is an enumeration and that's like a bunch of values that Vic
53:23
that's like a bunch of values that Vic
53:23
that's like a bunch of values that Vic that work for that slot for example on
53:28
that work for that slot for example on
53:28
that work for that slot for example on both Alexa and Google assistant they
53:31
both Alexa and Google assistant they
53:31
both Alexa and Google assistant they have native slots that are built in that
53:34
have native slots that are built in that
53:34
have native slots that are built in that are like movie names or the or
53:37
are like movie names or the or
53:37
are like movie names or the or sports team names things like that so
53:40
sports team names things like that so
53:40
sports team names things like that so you'll you'll see a slot that has the
53:43
you'll you'll see a slot that has the
53:43
you'll you'll see a slot that has the names of actors they you know things
53:45
names of actors they you know things
53:45
names of actors they you know things like that or you'll even see a first
53:46
like that or you'll even see a first
53:46
like that or you'll even see a first name slot and so what that is is that's
53:48
name slot and so what that is is that's
53:48
name slot and so what that is is that's just basically it's a giant enumeration
53:51
just basically it's a giant enumeration
53:51
just basically it's a giant enumeration which is pot all the possible values
53:53
which is pot all the possible values
53:53
which is pot all the possible values that could be for that slot and then
53:55
that could be for that slot and then
53:55
that could be for that slot and then what you do is you also define
53:57
what you do is you also define
53:57
what you do is you also define alternative ways to pronounce those
53:59
alternative ways to pronounce those
53:59
alternative ways to pronounce those names so for example for that for the
54:03
names so for example for that for the
54:03
names so for example for that for the Alexus scale I showed you for whetstone
54:04
Alexus scale I showed you for whetstone
54:04
Alexus scale I showed you for whetstone technologies I've created an enumeration
54:08
technologies I've created an enumeration
54:08
technologies I've created an enumeration called whetstone names that has the name
54:11
called whetstone names that has the name
54:11
called whetstone names that has the name whetstone technologies and sada bridge
54:15
whetstone technologies and sada bridge
54:15
whetstone technologies and sada bridge as to define Tut as to members of the
54:18
as to define Tut as to members of the
54:18
as to define Tut as to members of the enumeration and then under whetstone
54:21
enumeration and then under whetstone
54:21
enumeration and then under whetstone technologies I have multiple ways to
54:24
technologies I have multiple ways to
54:24
technologies I have multiple ways to specify whetstone which include you know
54:27
specify whetstone which include you know
54:27
specify whetstone which include you know just whetstone itself means whetstone
54:30
just whetstone itself means whetstone
54:30
just whetstone itself means whetstone technologies it also says whetstone I I
54:35
technologies it also says whetstone I I
54:35
technologies it also says whetstone I I noticed sometimes that the NLP would
54:38
noticed sometimes that the NLP would
54:38
noticed sometimes that the NLP would pick up what stone WH 80 stone so I went
54:42
pick up what stone WH 80 stone so I went
54:42
pick up what stone WH 80 stone so I went ahead and put that in there with sauna
54:46
ahead and put that in there with sauna
54:46
ahead and put that in there with sauna bridge I saw that often when it heard
54:48
bridge I saw that often when it heard
54:48
bridge I saw that often when it heard this it didn't hear the term sauna
54:50
this it didn't hear the term sauna
54:50
this it didn't hear the term sauna bridge because this speaks to its a
54:51
bridge because this speaks to its a
54:51
bridge because this speaks to its a brand name it's a name we made up so saw
54:54
brand name it's a name we made up so saw
54:54
brand name it's a name we made up so saw what it would actually here is sauna as
54:57
what it would actually here is sauna as
54:57
what it would actually here is sauna as in sau and a space bridge so I specify
55:01
in sau and a space bridge so I specify
55:01
in sau and a space bridge so I specify that as a synonym for sauna bridge and
55:03
that as a synonym for sauna bridge and
55:03
that as a synonym for sauna bridge and that gets us that actually getting you
55:06
that gets us that actually getting you
55:06
that gets us that actually getting you know that allows the that allows the
55:10
know that allows the that allows the
55:10
know that allows the that allows the virtual assistant gives it some hints to
55:12
virtual assistant gives it some hints to
55:12
virtual assistant gives it some hints to say oh yeah he said sauna bridge that's
55:15
say oh yeah he said sauna bridge that's
55:15
say oh yeah he said sauna bridge that's what I heard but what he really meant
55:16
what I heard but what he really meant
55:16
what I heard but what he really meant was the one word sauna bridge does that
55:19
was the one word sauna bridge does that
55:19
was the one word sauna bridge does that answer your question
55:20
answer your question
55:20
answer your question yep that's perfect that's perfect up one
55:24
yep that's perfect that's perfect up one
55:24
yep that's perfect that's perfect up one thing Sanjeev that you Oh even in the
55:26
thing Sanjeev that you Oh even in the
55:26
thing Sanjeev that you Oh even in the beginning of the slides you gave a demo
55:28
beginning of the slides you gave a demo
55:28
beginning of the slides you gave a demo for your Alexa skill right so is it
55:32
for your Alexa skill right so is it
55:32
for your Alexa skill right so is it possible to see now I have a dot with me
55:34
possible to see now I have a dot with me
55:34
possible to see now I have a dot with me right now okay
55:35
right now okay
55:35
right now okay grab a dot with me I'm in Asia now so
55:38
grab a dot with me I'm in Asia now so
55:38
grab a dot with me I'm in Asia now so how difference is the experience a
55:40
how difference is the experience a
55:40
how difference is the experience a person would be of using an Alexa from
55:43
person would be of using an Alexa from
55:43
person would be of using an Alexa from Asia or for a Europe not just LX I'd
55:46
Asia or for a Europe not just LX I'd
55:46
Asia or for a Europe not just LX I'd maybe Bigsby or a Siri how difference is
55:49
maybe Bigsby or a Siri how difference is
55:49
maybe Bigsby or a Siri how difference is the experience and how much
55:51
the experience and how much
55:51
the experience and how much so here here's the that the the short
55:56
so here here's the that the the short
55:56
so here here's the that the the short answer is the underlying the underlying
56:00
answer is the underlying the underlying
56:00
answer is the underlying the underlying application doesn't change so much
56:02
application doesn't change so much
56:02
application doesn't change so much because what you're getting is intense
56:04
because what you're getting is intense
56:04
because what you're getting is intense and all of that but what has to change
56:07
and all of that but what has to change
56:07
and all of that but what has to change for each one of those locales is the is
56:10
for each one of those locales is the is
56:10
for each one of those locales is the is the language model for example your ear
56:14
the language model for example your ear
56:14
the language model for example your ear in Asia
56:15
in Asia
56:15
in Asia I'm gonna guess you're in India and I'm
56:19
I'm gonna guess you're in India and I'm
56:19
I'm gonna guess you're in India and I'm just throwing that one out there you
56:22
just throwing that one out there you
56:22
just throwing that one out there you know so if I wanted to make my
56:24
know so if I wanted to make my
56:24
know so if I wanted to make my application also work in Hindi I'm gonna
56:28
application also work in Hindi I'm gonna
56:28
application also work in Hindi I'm gonna have to define a language model for
56:31
have to define a language model for
56:31
have to define a language model for Hindi right I could still have a yes
56:35
Hindi right I could still have a yes
56:35
Hindi right I could still have a yes intent but now it's gonna have to
56:38
intent but now it's gonna have to
56:38
intent but now it's gonna have to recognize ha right as the that I said
56:41
recognize ha right as the that I said
56:41
recognize ha right as the that I said yes or you know night that I said no
56:44
yes or you know night that I said no
56:44
yes or you know night that I said no it's going to have to understand that it
56:46
it's going to have to understand that it
56:46
it's going to have to understand that it wouldn't make sense in an English model
56:48
wouldn't make sense in an English model
56:48
wouldn't make sense in an English model but in a Hindi model it's gonna have to
56:51
but in a Hindi model it's gonna have to
56:51
but in a Hindi model it's gonna have to do that you know and then the you know
56:55
do that you know and then the you know
56:55
do that you know and then the you know you and and then you also have for
56:58
you and and then you also have for
56:58
you and and then you also have for example different flavor you know this
57:00
example different flavor you know this
57:00
example different flavor you know this go this always goes back to localization
57:02
go this always goes back to localization
57:02
go this always goes back to localization because you also have you not only have
57:05
because you also have you not only have
57:05
because you also have you not only have the language but then you have culture
57:06
the language but then you have culture
57:06
the language but then you have culture so for example with English if you just
57:09
so for example with English if you just
57:09
so for example with English if you just look at Alexa you have en - us you have
57:13
look at Alexa you have en - us you have
57:13
look at Alexa you have en - us you have en - can't see a for Canada en - GB for
57:18
en - can't see a for Canada en - GB for
57:18
en - can't see a for Canada en - GB for Great Britain Ian - I n for India and I
57:22
Great Britain Ian - I n for India and I
57:22
Great Britain Ian - I n for India and I tell you when I was doing some testing
57:25
tell you when I was doing some testing
57:25
tell you when I was doing some testing with one application we'd written for
57:28
with one application we'd written for
57:28
with one application we'd written for for en - I n it will wait look on the
57:33
for en - I n it will wait look on the
57:33
for en - I n it will wait look on the accent so you know being being being an
57:38
accent so you know being being being an
57:38
accent so you know being being being an american-born se myself right so you
57:42
american-born se myself right so you
57:42
american-born se myself right so you know you don't really hear an accent
57:44
know you don't really hear an accent
57:44
know you don't really hear an accent from me but you know the thing what I
57:46
from me but you know the thing what I
57:46
from me but you know the thing what I found was like so I'm used to saying
57:49
found was like so I'm used to saying
57:49
found was like so I'm used to saying stop stop I mean that's that's how you
57:51
stop stop I mean that's that's how you
57:51
stop stop I mean that's that's how you say it in the US right but when I switch
57:54
say it in the US right but when I switch
57:54
say it in the US right but when I switch to en - I n it would not recognize that
57:57
to en - I n it would not recognize that
57:57
to en - I n it would not recognize that most of the time when I started
57:59
most of the time when I started
57:59
most of the time when I started pronouncing it the way my parents would
58:01
pronouncing it the way my parents would
58:01
pronouncing it the way my parents would with an Indian accent stop
58:04
with an Indian accent stop
58:04
with an Indian accent stop that it started picking that up so you
58:07
that it started picking that up so you
58:08
that it started picking that up so you know that is a that is a you know that
58:12
know that is a that is a you know that
58:12
know that is a that is a you know that is that you know the the actual
58:14
is that you know the the actual
58:14
is that you know the the actual underlying language models are do pay
58:17
underlying language models are do pay
58:17
underlying language models are do pay attention to accent and that so it's you
58:20
attention to accent and that so it's you
58:20
attention to accent and that so it's you know it's a you know and so the idea is
58:22
know it's a you know and so the idea is
58:22
know it's a you know and so the idea is is when you create an application if you
58:25
is when you create an application if you
58:25
is when you create an application if you want it to cross languages you need to
58:28
want it to cross languages you need to
58:28
want it to cross languages you need to create a different model and publish for
58:30
create a different model and publish for
58:30
create a different model and publish for each one of those languages that's
58:33
each one of those languages that's
58:33
each one of those languages that's perfect I think one question that
58:35
perfect I think one question that
58:35
perfect I think one question that follows along with your explanation is
58:37
follows along with your explanation is
58:37
follows along with your explanation is asked by badonk with an says does it
58:40
asked by badonk with an says does it
58:40
asked by badonk with an says does it consider all the related word used in
58:43
consider all the related word used in
58:43
consider all the related word used in intent or only those specific words that
58:45
intent or only those specific words that
58:45
intent or only those specific words that we may have divine defined during the
58:47
we may have divine defined during the
58:47
we may have divine defined during the development so it it depends on the
58:51
development so it it depends on the
58:51
development so it it depends on the assistant and you know generally what
58:55
assistant and you know generally what
58:55
assistant and you know generally what you know this is where AI comes in you
58:58
you know this is where AI comes in you
58:58
you know this is where AI comes in you want to put as many utterances as you
59:00
want to put as many utterances as you
59:00
want to put as many utterances as you can but the thing is is that the AI
59:04
can but the thing is is that the AI
59:04
can but the thing is is that the AI behind the scenes if you give enough
59:08
behind the scenes if you give enough
59:08
behind the scenes if you give enough hints then the idea is is you know and
59:12
hints then the idea is is you know and
59:12
hints then the idea is is you know and again this is all going to vary from
59:13
again this is all going to vary from
59:13
again this is all going to vary from assistant to assistant but generally
59:16
assistant to assistant but generally
59:16
assistant to assistant but generally what you're trying to give them is
59:17
what you're trying to give them is
59:17
what you're trying to give them is enough hints of these are the different
59:20
enough hints of these are the different
59:20
enough hints of these are the different things that get to this utterance that
59:22
things that get to this utterance that
59:22
things that get to this utterance that the AI can then if something sounds like
59:25
the AI can then if something sounds like
59:25
the AI can then if something sounds like it should fall into this it'll then make
59:27
it should fall into this it'll then make
59:27
it should fall into this it'll then make the decision that this you know that you
59:30
the decision that this you know that you
59:30
the decision that this you know that you see theoretically don't need to specify
59:33
see theoretically don't need to specify
59:33
see theoretically don't need to specify everything but it's a general right now
59:37
everything but it's a general right now
59:37
everything but it's a general right now it's generally a good idea to give it as
59:39
it's generally a good idea to give it as
59:39
it's generally a good idea to give it as much as many hints as possible so that
59:42
much as many hints as possible so that
59:42
much as many hints as possible so that the AI can actually start to make some
59:45
the AI can actually start to make some
59:45
the AI can actually start to make some of it like I've you know I've had people
59:47
of it like I've you know I've had people
59:47
of it like I've you know I've had people say oh well I've specified two or three
59:50
say oh well I've specified two or three
59:50
say oh well I've specified two or three utterance is that enough no all right I
59:54
utterance is that enough no all right I
59:54
utterance is that enough no all right I gotta train them on own right yeah I
59:56
gotta train them on own right yeah I
59:56
gotta train them on own right yeah I mean I would say for the phone number
59:59
mean I would say for the phone number
59:59
mean I would say for the phone number for the phone number intent for the
1:00:02
for the phone number intent for the
1:00:02
for the phone number intent for the whetstone skill there's at least ten or
1:00:04
whetstone skill there's at least ten or
1:00:04
whetstone skill there's at least ten or fifteen uh turan sizes that I specified
1:00:07
fifteen uh turan sizes that I specified
1:00:07
fifteen uh turan sizes that I specified just for that alone yeah and and to
1:00:09
just for that alone yeah and and to
1:00:09
just for that alone yeah and and to answer that getting numbers is reliable
1:00:15
answer that getting numbers is reliable
1:00:15
answer that getting numbers is reliable so the
1:00:17
so the
1:00:17
so the assistants will here you know here you
1:00:21
assistants will here you know here you
1:00:21
assistants will here you know here you say zero one two three four five six
1:00:23
say zero one two three four five six
1:00:23
say zero one two three four five six seven eight nine
1:00:24
seven eight nine
1:00:24
seven eight nine they'll understand that rather well but
1:00:28
they'll understand that rather well but
1:00:28
they'll understand that rather well but spelling your name spelling an email is
1:00:32
spelling your name spelling an email is
1:00:32
spelling your name spelling an email is not not reliable so I'll take my last
1:00:36
not not reliable so I'll take my last
1:00:36
not not reliable so I'll take my last name for example I WASD it's pronounced
1:00:39
name for example I WASD it's pronounced
1:00:39
name for example I WASD it's pronounced II was you can guess what the chances
1:00:42
II was you can guess what the chances
1:00:42
II was you can guess what the chances are that if I say II was to Alexa that
1:00:45
are that if I say II was to Alexa that
1:00:45
are that if I say II was to Alexa that I'll get the spelling correct right and
1:00:48
I'll get the spelling correct right and
1:00:48
I'll get the spelling correct right and and Sanjeev it may or may not get that
1:00:52
and Sanjeev it may or may not get that
1:00:52
and Sanjeev it may or may not get that correct even though that that's a little
1:00:53
correct even though that that's a little
1:00:53
correct even though that that's a little bit more clearly pronounced sonj attempt
1:00:57
bit more clearly pronounced sonj attempt
1:00:57
bit more clearly pronounced sonj attempt we we did an experiment where we
1:00:59
we we did an experiment where we
1:01:00
we we did an experiment where we attempted to speak our own names as a as
1:01:06
attempted to speak our own names as a as
1:01:06
attempted to speak our own names as a as an intent with the letters as individual
1:01:08
an intent with the letters as individual
1:01:09
an intent with the letters as individual slot values to both Alexa and Google and
1:01:13
slot values to both Alexa and Google and
1:01:13
slot values to both Alexa and Google and met with similar experiences on both
1:01:18
met with similar experiences on both
1:01:18
met with similar experiences on both platforms the Z in my name came in as
1:01:25
platforms the Z in my name came in as
1:01:25
platforms the Z in my name came in as sometimes it heard B sometimes it heard
1:01:28
sometimes it heard B sometimes it heard
1:01:28
sometimes it heard B sometimes it heard V Sandra's name when we spelled sa n je
1:01:34
V Sandra's name when we spelled sa n je
1:01:34
V Sandra's name when we spelled sa n je e V it would skip the second E
1:01:37
e V it would skip the second E
1:01:37
e V it would skip the second E you know 19 times out 20 it did not hear
1:01:40
you know 19 times out 20 it did not hear
1:01:40
you know 19 times out 20 it did not hear the second e so you may think that you
1:01:45
the second e so you may think that you
1:01:45
the second e so you may think that you know you can get around some of this by
1:01:48
know you can get around some of this by
1:01:48
know you can get around some of this by having the user spell the words spell
1:01:51
having the user spell the words spell
1:01:51
having the user spell the words spell their email that's that's not going to
1:01:53
their email that's that's not going to
1:01:53
their email that's that's not going to work you do have a better chance at
1:01:56
work you do have a better chance at
1:01:56
work you do have a better chance at having the user speak a number or de
1:02:00
having the user speak a number or de
1:02:00
having the user speak a number or de Noli so if you say you know 1 2 3 you
1:02:03
Noli so if you say you know 1 2 3 you
1:02:03
Noli so if you say you know 1 2 3 you can bet that you can get more reliably
1:02:06
can bet that you can get more reliably
1:02:06
can bet that you can get more reliably yeah having said that yangsun design you
1:02:10
yeah having said that yangsun design you
1:02:10
yeah having said that yangsun design you know just I mean again that speaks to
1:02:12
know just I mean again that speaks to
1:02:12
know just I mean again that speaks to training you have to have enough
1:02:14
training you have to have enough
1:02:14
training you have to have enough training in that model to get there and
1:02:17
training in that model to get there and
1:02:17
training in that model to get there and having in a sit you know if you just
1:02:20
having in a sit you know if you just
1:02:20
having in a sit you know if you just have 26 letters how does it you know
1:02:23
have 26 letters how does it you know
1:02:23
have 26 letters how does it you know there's not enough training for it to
1:02:24
there's not enough training for it to
1:02:24
there's not enough training for it to recognize the difference between C and Z
1:02:27
recognize the difference between C and Z
1:02:27
recognize the difference between C and Z or D and T yeah you know it
1:02:30
or D and T yeah you know it
1:02:30
or D and T yeah you know it you know so that again that that's a
1:02:32
you know so that again that that's a
1:02:32
you know so that again that that's a good example I think of why training and
1:02:35
good example I think of why training and
1:02:35
good example I think of why training and providing enough examples is important
1:02:38
providing enough examples is important
1:02:38
providing enough examples is important because a lot of this is around pattern
1:02:40
because a lot of this is around pattern
1:02:40
because a lot of this is around pattern recognition yep yep oh one thing to note
1:02:44
recognition yep yep oh one thing to note
1:02:44
recognition yep yep oh one thing to note here sanjeev is you might get some SMS
1:02:46
here sanjeev is you might get some SMS
1:02:46
here sanjeev is you might get some SMS and calls today because he's a shady a
1:02:48
and calls today because he's a shady a
1:02:48
and calls today because he's a shady a contact number on the live show yeah
1:02:51
contact number on the live show yeah
1:02:51
contact number on the live show yeah right so yeah adding on to I think we
1:02:54
right so yeah adding on to I think we
1:02:54
right so yeah adding on to I think we have reached one upper there couple of
1:02:56
have reached one upper there couple of
1:02:56
have reached one upper there couple of questions you would like to before just
1:02:57
questions you would like to before just
1:02:57
questions you would like to before just wrap it up one thing that I would like
1:03:00
wrap it up one thing that I would like
1:03:00
wrap it up one thing that I would like to ask is what is the impact of IOT in
1:03:04
to ask is what is the impact of IOT in
1:03:04
to ask is what is the impact of IOT in the on voice of voice technology on IOT
1:03:06
the on voice of voice technology on IOT
1:03:06
the on voice of voice technology on IOT we see we can connect to appliances that
1:03:09
we see we can connect to appliances that
1:03:09
we see we can connect to appliances that the smart cities they've added digital
1:03:12
the smart cities they've added digital
1:03:12
the smart cities they've added digital you know projects coming so how do we
1:03:15
you know projects coming so how do we
1:03:15
you know projects coming so how do we see voice technology moving into IOD and
1:03:17
see voice technology moving into IOD and
1:03:17
see voice technology moving into IOD and what is what's next in this yeah so I I
1:03:23
what is what's next in this yeah so I I
1:03:23
what is what's next in this yeah so I I can give a I can tell you about my own
1:03:28
can give a I can tell you about my own
1:03:28
can give a I can tell you about my own experience with wiring up at home you
1:03:32
experience with wiring up at home you
1:03:32
experience with wiring up at home you know Wi-Fi enabled lights like the
1:03:35
know Wi-Fi enabled lights like the
1:03:35
know Wi-Fi enabled lights like the Phillips hue lights have become rather
1:03:36
Phillips hue lights have become rather
1:03:36
Phillips hue lights have become rather popular and at home you know I have my
1:03:39
popular and at home you know I have my
1:03:39
popular and at home you know I have my Wi-Fi lights connected to my Alexa
1:03:43
Wi-Fi lights connected to my Alexa
1:03:43
Wi-Fi lights connected to my Alexa device and so I can enable simple
1:03:47
device and so I can enable simple
1:03:47
device and so I can enable simple scenarios like changing turning the
1:03:49
scenarios like changing turning the
1:03:49
scenarios like changing turning the light on or off by voice setting
1:03:52
light on or off by voice setting
1:03:52
light on or off by voice setting routines so I can tell Alexa you know
1:03:55
routines so I can tell Alexa you know
1:03:55
routines so I can tell Alexa you know I'm I'm going to bed now and it'll dim
1:03:57
I'm I'm going to bed now and it'll dim
1:03:57
I'm I'm going to bed now and it'll dim the lights downstairs I can do a fun one
1:04:01
the lights downstairs I can do a fun one
1:04:01
the lights downstairs I can do a fun one which is great party trick where I can
1:04:02
which is great party trick where I can
1:04:02
which is great party trick where I can say you know Alexa you know till if this
1:04:06
say you know Alexa you know till if this
1:04:06
say you know Alexa you know till if this then that red alert and you'll hear
1:04:10
then that red alert and you'll hear
1:04:10
then that red alert and you'll hear Alexa play you know the Star Trek
1:04:11
Alexa play you know the Star Trek
1:04:11
Alexa play you know the Star Trek Klaxons like the ship is about to
1:04:13
Klaxons like the ship is about to
1:04:13
Klaxons like the ship is about to self-destruct and my lens will flash red
1:04:15
self-destruct and my lens will flash red
1:04:15
self-destruct and my lens will flash red and it would that drives my wife crazy
1:04:18
and it would that drives my wife crazy
1:04:18
and it would that drives my wife crazy but I love it in you can also see IOT
1:04:25
but I love it in you can also see IOT
1:04:25
but I love it in you can also see IOT enabled scenarios in business functions
1:04:28
enabled scenarios in business functions
1:04:28
enabled scenarios in business functions so to make that a little more concrete I
1:04:32
so to make that a little more concrete I
1:04:32
so to make that a little more concrete I have a friend who runs a business that
1:04:37
have a friend who runs a business that
1:04:37
have a friend who runs a business that distributes and configures equipment for
1:04:42
distributes and configures equipment for
1:04:42
distributes and configures equipment for performing
1:04:44
performing
1:04:44
performing intra vitro fertilization treatments for
1:04:48
intra vitro fertilization treatments for
1:04:48
intra vitro fertilization treatments for people that you know need a little bit
1:04:50
people that you know need a little bit
1:04:50
people that you know need a little bit of help getting pregnant and the
1:04:53
of help getting pregnant and the
1:04:53
of help getting pregnant and the clinicians that operate the equipment
1:04:57
clinicians that operate the equipment
1:04:57
clinicians that operate the equipment often have their hands full as you can
1:05:00
often have their hands full as you can
1:05:00
often have their hands full as you can imagine that devices are very sensitive
1:05:02
imagine that devices are very sensitive
1:05:02
imagine that devices are very sensitive and we have been in discussion in
1:05:06
and we have been in discussion in
1:05:06
and we have been in discussion in creating voice-enabled scenario which
1:05:08
creating voice-enabled scenario which
1:05:08
creating voice-enabled scenario which would be more of an OEM and delivery
1:05:10
would be more of an OEM and delivery
1:05:10
would be more of an OEM and delivery where the the clinician could ask for
1:05:16
where the the clinician could ask for
1:05:16
where the the clinician could ask for the microscope to zoom in or zoom out
1:05:18
the microscope to zoom in or zoom out
1:05:18
the microscope to zoom in or zoom out while their hands are occupied on other
1:05:20
while their hands are occupied on other
1:05:20
while their hands are occupied on other diets I saw a question earlier about
1:05:24
diets I saw a question earlier about
1:05:24
diets I saw a question earlier about what applications this has to - defense
1:05:27
what applications this has to - defense
1:05:27
what applications this has to - defense - I assume you buy it by defense you
1:05:30
- I assume you buy it by defense you
1:05:30
- I assume you buy it by defense you mean military applications and you can
1:05:33
mean military applications and you can
1:05:33
mean military applications and you can apply that same kind of use case
1:05:37
apply that same kind of use case
1:05:37
apply that same kind of use case scenario where you have an individual
1:05:38
scenario where you have an individual
1:05:38
scenario where you have an individual whose hands are occupied but they you
1:05:43
whose hands are occupied but they you
1:05:43
whose hands are occupied but they you know ideally you know they if they could
1:05:45
know ideally you know they if they could
1:05:45
know ideally you know they if they could grow a third arm they would it's in
1:05:47
grow a third arm they would it's in
1:05:47
grow a third arm they would it's in those scenarios where having voice
1:05:50
those scenarios where having voice
1:05:50
those scenarios where having voice enabled devices in IOT integration would
1:05:53
enabled devices in IOT integration would
1:05:53
enabled devices in IOT integration would make the job so much easier but that's
1:05:57
make the job so much easier but that's
1:05:57
make the job so much easier but that's perfect I think we'll take oh maybe this
1:06:00
perfect I think we'll take oh maybe this
1:06:00
perfect I think we'll take oh maybe this may be the second loss that we take my H
1:06:02
may be the second loss that we take my H
1:06:02
may be the second loss that we take my H bene well goes in and praises that is it
1:06:05
bene well goes in and praises that is it
1:06:05
bene well goes in and praises that is it amazing session you guys are having what
1:06:07
amazing session you guys are having what
1:06:07
amazing session you guys are having what about the growth or market do you see
1:06:10
about the growth or market do you see
1:06:10
about the growth or market do you see for voice technology in coming years I
1:06:13
for voice technology in coming years I
1:06:13
for voice technology in coming years I you know yeah I think that that really
1:06:16
you know yeah I think that that really
1:06:16
you know yeah I think that that really just speaks to you know the the the
1:06:19
just speaks to you know the the the
1:06:19
just speaks to you know the the the that's the that's the question everybody
1:06:22
that's the that's the question everybody
1:06:22
that's the that's the question everybody you know you you keep thinking okay it's
1:06:25
you know you you keep thinking okay it's
1:06:25
you know you you keep thinking okay it's gonna happen this year it's gonna blow
1:06:27
gonna happen this year it's gonna blow
1:06:27
gonna happen this year it's gonna blow up because you have this sits you have
1:06:29
up because you have this sits you have
1:06:29
up because you have this sits you have this situation where the consumer
1:06:31
this situation where the consumer
1:06:31
this situation where the consumer adoption is just ridiculous
1:06:39
that's stage this is mainstream
1:06:41
that's stage this is mainstream
1:06:41
that's stage this is mainstream technology so you know from a consumers
1:06:44
technology so you know from a consumers
1:06:44
technology so you know from a consumers perspective it's out there the you know
1:06:47
perspective it's out there the you know
1:06:47
perspective it's out there the you know the the thing that's really kind of
1:06:49
the the thing that's really kind of
1:06:49
the the thing that's really kind of gating things is from a lot of business
1:06:51
gating things is from a lot of business
1:06:51
gating things is from a lot of business perspective it's still in the it's still
1:06:55
perspective it's still in the it's still
1:06:55
perspective it's still in the it's still in
1:06:57
in
1:06:57
in you know in the early early stages of
1:06:59
you know in the early early stages of
1:06:59
you know in the early early stages of what can we do with this and you know so
1:07:02
what can we do with this and you know so
1:07:02
what can we do with this and you know so there's a lot of experimentation going
1:07:03
there's a lot of experimentation going
1:07:03
there's a lot of experimentation going on the killer app has not appeared yet
1:07:08
on the killer app has not appeared yet
1:07:08
on the killer app has not appeared yet that says okay blow up but you know when
1:07:13
that says okay blow up but you know when
1:07:13
that says okay blow up but you know when you look at everything it does point to
1:07:15
you look at everything it does point to
1:07:15
you look at everything it does point to again it's what I said before it's not
1:07:17
again it's what I said before it's not
1:07:17
again it's what I said before it's not an if it's a win oh yeah yeah but to
1:07:22
an if it's a win oh yeah yeah but to
1:07:22
an if it's a win oh yeah yeah but to build on that I think we're going to see
1:07:25
build on that I think we're going to see
1:07:25
build on that I think we're going to see the current pandemic informing the
1:07:27
the current pandemic informing the
1:07:27
the current pandemic informing the direction of where voice technologies
1:07:29
direction of where voice technologies
1:07:29
direction of where voice technologies applied in the the next 12 to 18 months
1:07:33
applied in the the next 12 to 18 months
1:07:33
applied in the the next 12 to 18 months one good example and I'll ask assignment
1:07:37
one good example and I'll ask assignment
1:07:37
one good example and I'll ask assignment to shared this link I just put in
1:07:38
to shared this link I just put in
1:07:38
to shared this link I just put in private chat just on on CNBC just on
1:07:44
private chat just on on CNBC just on
1:07:44
private chat just on on CNBC just on looking at the date just on the 12th
1:07:45
looking at the date just on the 12th
1:07:45
looking at the date just on the 12th just two days ago on mainstream
1:07:49
just two days ago on mainstream
1:07:49
just two days ago on mainstream television there was a scenario laid out
1:07:53
television there was a scenario laid out
1:07:53
television there was a scenario laid out where as we continue to go shopping and
1:07:58
where as we continue to go shopping and
1:07:58
where as we continue to go shopping and retail shopping in the colder months
1:08:01
retail shopping in the colder months
1:08:01
retail shopping in the colder months you're not going to be waiting in line
1:08:03
you're not going to be waiting in line
1:08:03
you're not going to be waiting in line to get into the store in the scenario
1:08:06
to get into the store in the scenario
1:08:06
to get into the store in the scenario laid out in this article they talked
1:08:08
laid out in this article they talked
1:08:08
laid out in this article they talked about sending push notifications that
1:08:10
about sending push notifications that
1:08:10
about sending push notifications that enable kind of a deli counter
1:08:12
enable kind of a deli counter
1:08:12
enable kind of a deli counter kind of interaction where when you
1:08:15
kind of interaction where when you
1:08:15
kind of interaction where when you arrive at the store you may or may not
1:08:17
arrive at the store you may or may not
1:08:17
arrive at the store you may or may not be able to go in right away there may be
1:08:19
be able to go in right away there may be
1:08:19
be able to go in right away there may be a queue in the cold months you don't
1:08:21
a queue in the cold months you don't
1:08:21
a queue in the cold months you don't want to wait outside you want to wait in
1:08:22
want to wait outside you want to wait in
1:08:22
want to wait outside you want to wait in your car and so you would get a push
1:08:25
your car and so you would get a push
1:08:25
your car and so you would get a push notification on your phone that would be
1:08:30
notification on your phone that would be
1:08:30
notification on your phone that would be say the Walmart speaker experience where
1:08:37
say the Walmart speaker experience where
1:08:37
say the Walmart speaker experience where if that would be integrated with your
1:08:38
if that would be integrated with your
1:08:38
if that would be integrated with your phone with your car to inform you to
1:08:40
phone with your car to inform you to
1:08:40
phone with your car to inform you to tell you what the queue time is what the
1:08:44
tell you what the queue time is what the
1:08:44
tell you what the queue time is what the expected wait time is so I think we're
1:08:47
expected wait time is so I think we're
1:08:47
expected wait time is so I think we're going to see voice experiences in novel
1:08:51
going to see voice experiences in novel
1:08:51
going to see voice experiences in novel ways that we haven't seen before as it
1:08:54
ways that we haven't seen before as it
1:08:54
ways that we haven't seen before as it begins to address the pandemic and
1:08:57
begins to address the pandemic and
1:08:57
begins to address the pandemic and address reducing direct contact with
1:09:02
address reducing direct contact with
1:09:02
address reducing direct contact with surfaces as well as encouraging social
1:09:05
surfaces as well as encouraging social
1:09:05
surfaces as well as encouraging social distancing yeah that I think social
1:09:09
distancing yeah that I think social
1:09:09
distancing yeah that I think social distancing and
1:09:10
distancing and
1:09:10
distancing and working from home is gonna be the new
1:09:12
working from home is gonna be the new
1:09:12
working from home is gonna be the new normal now adding to it we do have a
1:09:15
normal now adding to it we do have a
1:09:15
normal now adding to it we do have a voice technology coming to self-driving
1:09:17
voice technology coming to self-driving
1:09:17
voice technology coming to self-driving cars I mean we have seen in Tesla just
1:09:20
cars I mean we have seen in Tesla just
1:09:20
cars I mean we have seen in Tesla just give a voice come on hey take me to the
1:09:22
give a voice come on hey take me to the
1:09:22
give a voice come on hey take me to the home and the car will drive and take it
1:09:24
home and the car will drive and take it
1:09:24
home and the car will drive and take it you may be a workplace or to your home
1:09:27
you may be a workplace or to your home
1:09:27
you may be a workplace or to your home one last question that I I keep on
1:09:30
one last question that I I keep on
1:09:30
one last question that I I keep on saying it's the last question but it
1:09:31
saying it's the last question but it
1:09:31
saying it's the last question but it keeps on coming Sandeep go ahead ahead
1:09:34
keeps on coming Sandeep go ahead ahead
1:09:34
keeps on coming Sandeep go ahead ahead and ask for an opinion from both of you
1:09:37
and ask for an opinion from both of you
1:09:37
and ask for an opinion from both of you is that will there be a time when these
1:09:39
is that will there be a time when these
1:09:39
is that will there be a time when these three big companies like Amazon Google
1:09:41
three big companies like Amazon Google
1:09:41
three big companies like Amazon Google or Apple will share the matrices on how
1:09:44
or Apple will share the matrices on how
1:09:44
or Apple will share the matrices on how they work and how close they're with any
1:09:47
they work and how close they're with any
1:09:47
they work and how close they're with any of the intent or the that you provide so
1:09:52
of the intent or the that you provide so
1:09:52
of the intent or the that you provide so you know ulster so as far as sharing
1:09:57
you know ulster so as far as sharing
1:09:57
you know ulster so as far as sharing metrics you know this is this is one of
1:10:00
metrics you know this is this is one of
1:10:00
metrics you know this is this is one of those things where right now it just
1:10:03
those things where right now it just
1:10:03
those things where right now it just doesn't happen you have companies that
1:10:06
doesn't happen you have companies that
1:10:06
doesn't happen you have companies that like ours that right you know that
1:10:08
like ours that right you know that
1:10:08
like ours that right you know that effectively right platforms where we're
1:10:11
effectively right platforms where we're
1:10:11
effectively right platforms where we're getting everything coming through the
1:10:13
getting everything coming through the
1:10:13
getting everything coming through the same you know since it's coming through
1:10:15
same you know since it's coming through
1:10:15
same you know since it's coming through the same platform we've effectively
1:10:17
the same platform we've effectively
1:10:17
the same platform we've effectively normalized those metrics you know the
1:10:20
normalized those metrics you know the
1:10:20
normalized those metrics you know the real question I think in John maybe you
1:10:21
real question I think in John maybe you
1:10:21
real question I think in John maybe you can speak more to this is like with that
1:10:23
can speak more to this is like with that
1:10:23
can speak more to this is like with that that seems to me to be one of the things
1:10:26
that seems to me to be one of the things
1:10:26
that seems to me to be one of the things that they're trying to get to with the
1:10:27
that they're trying to get to with the
1:10:27
that they're trying to get to with the open voice Network is a standardization
1:10:31
open voice Network is a standardization
1:10:31
open voice Network is a standardization but I don't know how far you know we're
1:10:33
but I don't know how far you know we're
1:10:33
but I don't know how far you know we're I don't I don't know where that sitting
1:10:36
I don't I don't know where that sitting
1:10:36
I don't I don't know where that sitting on the priority list for them well one
1:10:39
on the priority list for them well one
1:10:39
on the priority list for them well one one kind of question that we've done
1:10:41
one kind of question that we've done
1:10:41
one kind of question that we've done from customers is when considering the
1:10:45
from customers is when considering the
1:10:45
from customers is when considering the rollout of a skill the one customer
1:10:49
rollout of a skill the one customer
1:10:49
rollout of a skill the one customer asked us could you give us a breakdown
1:10:50
asked us could you give us a breakdown
1:10:50
asked us could you give us a breakdown by zip code what the adoption rate is of
1:10:54
by zip code what the adoption rate is of
1:10:54
by zip code what the adoption rate is of you know smart speakers and I can't you
1:10:57
you know smart speakers and I can't you
1:10:57
you know smart speakers and I can't you know Alexa and Google does not release
1:10:59
know Alexa and Google does not release
1:10:59
know Alexa and Google does not release that kind of demographic information
1:11:00
that kind of demographic information
1:11:00
that kind of demographic information they keep that very close to the best
1:11:03
they keep that very close to the best
1:11:03
they keep that very close to the best the best answer for adoption rates for
1:11:07
the best answer for adoption rates for
1:11:07
the best answer for adoption rates for the number of devices throughout a
1:11:12
the number of devices throughout a
1:11:12
the number of devices throughout a country or region is voice plant may I
1:11:16
country or region is voice plant may I
1:11:16
country or region is voice plant may I they do a good job of aggregating the
1:11:19
they do a good job of aggregating the
1:11:19
they do a good job of aggregating the adoption rates of these speakers the the
1:11:22
adoption rates of these speakers the the
1:11:22
adoption rates of these speakers the the use and demographics
1:11:24
use and demographics
1:11:24
use and demographics of these speakers and they give a good
1:11:26
of these speakers and they give a good
1:11:26
of these speakers and they give a good breakdown of how how people tend to
1:11:29
breakdown of how how people tend to
1:11:29
breakdown of how how people tend to interact you know what percentage of
1:11:31
interact you know what percentage of
1:11:31
interact you know what percentage of people tend to use them for timers for
1:11:33
people tend to use them for timers for
1:11:33
people tend to use them for timers for music for third-party applications etc
1:11:35
music for third-party applications etc
1:11:35
music for third-party applications etc so the answer to that question I think
1:11:38
so the answer to that question I think
1:11:38
so the answer to that question I think right now today is go to Boise bought a
1:11:42
right now today is go to Boise bought a
1:11:42
right now today is go to Boise bought a I and as for when these when you know a
1:11:49
I and as for when these when you know a
1:11:49
I and as for when these when you know a lecture Google would release their own
1:11:52
lecture Google would release their own
1:11:52
lecture Google would release their own adoption rates and you know more
1:11:55
adoption rates and you know more
1:11:55
adoption rates and you know more fine-grain detail information I have no
1:11:56
fine-grain detail information I have no
1:11:56
fine-grain detail information I have no idea I wouldn't hold my breath I think
1:12:02
idea I wouldn't hold my breath I think
1:12:02
idea I wouldn't hold my breath I think that well answered I think we have
1:12:05
that well answered I think we have
1:12:05
that well answered I think we have reached a time now I think right John at
1:12:07
reached a time now I think right John at
1:12:07
reached a time now I think right John at sanjeev you may have some other work to
1:12:10
sanjeev you may have some other work to
1:12:10
sanjeev you may have some other work to but i still see many people have joined
1:12:12
but i still see many people have joined
1:12:12
but i still see many people have joined us they're still joining us new people
1:12:14
us they're still joining us new people
1:12:14
us they're still joining us new people keep on joining so that i think indeed
1:12:16
keep on joining so that i think indeed
1:12:16
keep on joining so that i think indeed this session is really really amazing
1:12:18
this session is really really amazing
1:12:18
this session is really really amazing pub the moment this Q&A started there
1:12:20
pub the moment this Q&A started there
1:12:20
pub the moment this Q&A started there was a great spike of people joining us
1:12:23
was a great spike of people joining us
1:12:23
was a great spike of people joining us for the Q&A so of course it was a really
1:12:26
for the Q&A so of course it was a really
1:12:26
for the Q&A so of course it was a really interesting session and anything last
1:12:28
interesting session and anything last
1:12:28
interesting session and anything last words or tips that you both guys want to
1:12:32
words or tips that you both guys want to
1:12:32
words or tips that you both guys want to give to the developers coming from any
1:12:33
give to the developers coming from any
1:12:33
give to the developers coming from any of the background and if they want to
1:12:35
of the background and if they want to
1:12:35
of the background and if they want to move into this field any protea because
1:12:37
move into this field any protea because
1:12:37
move into this field any protea because you guys have been working on this for a
1:12:39
you guys have been working on this for a
1:12:39
you guys have been working on this for a pretty long time what are the tips that
1:12:40
pretty long time what are the tips that
1:12:40
pretty long time what are the tips that you would like to give for people who
1:12:42
you would like to give for people who
1:12:42
you would like to give for people who want to move into this voice technology
1:12:43
want to move into this voice technology
1:12:43
want to move into this voice technology and anyone from you can start so i want
1:12:47
and anyone from you can start so i want
1:12:47
and anyone from you can start so i want to take that one yeah yeah yeah
1:12:49
to take that one yeah yeah yeah
1:12:49
to take that one yeah yeah yeah you know honestly it's you know one of
1:12:52
you know honestly it's you know one of
1:12:52
you know honestly it's you know one of the things I've always found just you
1:12:54
the things I've always found just you
1:12:54
the things I've always found just you know I've been I've been I've been
1:12:56
know I've been I've been I've been
1:12:56
know I've been I've been I've been developing software for you know as I
1:12:59
developing software for you know as I
1:12:59
developing software for you know as I said at the beginning like 25 30 years
1:13:01
said at the beginning like 25 30 years
1:13:01
said at the beginning like 25 30 years just go do it you know they're all this
1:13:04
just go do it you know they're all this
1:13:04
just go do it you know they're all this stuff like doing an Alexis skill there
1:13:07
stuff like doing an Alexis skill there
1:13:07
stuff like doing an Alexis skill there are plenty of tutorials on how to do
1:13:09
are plenty of tutorials on how to do
1:13:09
are plenty of tutorials on how to do this
1:13:10
this
1:13:10
this amazon has create you know has a lot of
1:13:13
amazon has create you know has a lot of
1:13:13
amazon has create you know has a lot of nice libraries that make it very easy to
1:13:16
nice libraries that make it very easy to
1:13:16
nice libraries that make it very easy to just stand something up with nodejs and
1:13:20
just stand something up with nodejs and
1:13:20
just stand something up with nodejs and again with google there's plenty of
1:13:22
again with google there's plenty of
1:13:22
again with google there's plenty of examples on how to do this it's you know
1:13:25
examples on how to do this it's you know
1:13:25
examples on how to do this it's you know you don't even have to have an Alexa
1:13:27
you don't even have to have an Alexa
1:13:27
you don't even have to have an Alexa device to you know or or a Google
1:13:29
device to you know or or a Google
1:13:29
device to you know or or a Google assistant device because they have test
1:13:31
assistant device because they have test
1:13:31
assistant device because they have test harnesses so you can actually just go
1:13:34
harnesses so you can actually just go
1:13:34
harnesses so you can actually just go ahead stand something up and start
1:13:36
ahead stand something up and start
1:13:36
ahead stand something up and start playing or Pocono
1:13:37
playing or Pocono
1:13:37
playing or Pocono around with it I'm you know it's very
1:13:40
around with it I'm you know it's very
1:13:40
around with it I'm you know it's very very it's it's literally just go to the
1:13:43
very it's it's literally just go to the
1:13:43
very it's it's literally just go to the site and get started it's just like any
1:13:46
site and get started it's just like any
1:13:46
site and get started it's just like any other type of technology but I would
1:13:48
other type of technology but I would
1:13:48
other type of technology but I would answer that and say make sure you have a
1:13:49
answer that and say make sure you have a
1:13:49
answer that and say make sure you have a when you're developing a voice ID make
1:13:51
when you're developing a voice ID make
1:13:51
when you're developing a voice ID make sure you have a goal in mind I you know
1:13:55
sure you have a goal in mind I you know
1:13:55
sure you have a goal in mind I you know everybody learns differently and what
1:13:57
everybody learns differently and what
1:13:57
everybody learns differently and what seems to work for me when I'm trying to
1:13:58
seems to work for me when I'm trying to
1:13:58
seems to work for me when I'm trying to pick up a new new programming language
1:14:00
pick up a new new programming language
1:14:00
pick up a new new programming language is to Senegal decide what skill what
1:14:05
is to Senegal decide what skill what
1:14:05
is to Senegal decide what skill what experience you want to produce and let
1:14:09
experience you want to produce and let
1:14:09
experience you want to produce and let that inform and drive what you develop
1:14:12
that inform and drive what you develop
1:14:12
that inform and drive what you develop and and the skill that you build you
1:14:15
and and the skill that you build you
1:14:15
and and the skill that you build you know and as for you know I I do see the
1:14:19
know and as for you know I I do see the
1:14:19
know and as for you know I I do see the question on having just completed
1:14:22
question on having just completed
1:14:22
question on having just completed bachelors how do you get into the field
1:14:24
bachelors how do you get into the field
1:14:24
bachelors how do you get into the field of voice technology well I would say yes
1:14:26
of voice technology well I would say yes
1:14:26
of voice technology well I would say yes release release a sample skill that is a
1:14:32
release release a sample skill that is a
1:14:32
release release a sample skill that is a bit more feature-rich does something
1:14:34
bit more feature-rich does something
1:14:34
bit more feature-rich does something that somebody else hasn't done yet
1:14:36
that somebody else hasn't done yet
1:14:36
that somebody else hasn't done yet ideally you know make it multimodal make
1:14:39
ideally you know make it multimodal make
1:14:39
ideally you know make it multimodal make it also interact with the echo show you
1:14:43
it also interact with the echo show you
1:14:43
it also interact with the echo show you know show off
1:14:44
know show off
1:14:44
know show off and use that as a calling card for for
1:14:49
and use that as a calling card for for
1:14:49
and use that as a calling card for for applying to companies that are using
1:14:53
applying to companies that are using
1:14:53
applying to companies that are using voice technology specializing in voice
1:14:54
voice technology specializing in voice
1:14:54
voice technology specializing in voice technology let that be your shining mark
1:14:59
technology let that be your shining mark
1:14:59
technology let that be your shining mark on your resume
1:15:02
Sanji you would like to add something
1:15:04
Sanji you would like to add something
1:15:04
Sanji you would like to add something okay yeah yeah yeah I gave my two cents
1:15:10
okay yeah yeah yeah I gave my two cents
1:15:10
okay yeah yeah yeah I gave my two cents okay I think I've also showed a link
1:15:13
okay I think I've also showed a link
1:15:13
okay I think I've also showed a link that an article that John wrote on C
1:15:15
that an article that John wrote on C
1:15:15
that an article that John wrote on C shop corner some time back on how you
1:15:17
shop corner some time back on how you
1:15:17
shop corner some time back on how you can use Alexa without having an echo
1:15:19
can use Alexa without having an echo
1:15:19
can use Alexa without having an echo device so that is an article that John
1:15:21
device so that is an article that John
1:15:21
device so that is an article that John wrote and I have shared that in the
1:15:23
wrote and I have shared that in the
1:15:23
wrote and I have shared that in the comment section so that was a really
1:15:26
comment section so that was a really
1:15:26
comment section so that was a really great session Sanjeev encho on a lot to
1:15:28
great session Sanjeev encho on a lot to
1:15:28
great session Sanjeev encho on a lot to be taken away from just around 62 or 70
1:15:31
be taken away from just around 62 or 70
1:15:31
be taken away from just around 62 or 70 minutes of we really enjoyed it and on
1:15:35
minutes of we really enjoyed it and on
1:15:35
minutes of we really enjoyed it and on behalf of entire C shop Corner community
1:15:38
behalf of entire C shop Corner community
1:15:38
behalf of entire C shop Corner community and its millions of users I would like
1:15:40
and its millions of users I would like
1:15:40
and its millions of users I would like to thank you for your valuable time and
1:15:43
to thank you for your valuable time and
1:15:43
to thank you for your valuable time and your contributions towards the community
1:15:44
your contributions towards the community
1:15:44
your contributions towards the community thank you thank you so much once again
1:15:46
thank you thank you so much once again
1:15:46
thank you thank you so much once again and we would love to have you back on
1:15:48
and we would love to have you back on
1:15:48
and we would love to have you back on our show yeah was our pleasure I
1:15:51
our show yeah was our pleasure I
1:15:51
our show yeah was our pleasure I talk about this stuff at length and you
1:15:53
talk about this stuff at length and you
1:15:53
talk about this stuff at length and you can find me online at at John I 73 on on
1:15:58
can find me online at at John I 73 on on
1:15:58
can find me online at at John I 73 on on Twitter and of course I'm I've written a
1:16:01
Twitter and of course I'm I've written a
1:16:01
Twitter and of course I'm I've written a few articles on C sharp corner and you
1:16:03
few articles on C sharp corner and you
1:16:03
few articles on C sharp corner and you can find me there so thanks I really
1:16:05
can find me there so thanks I really
1:16:05
can find me there so thanks I really really appreciate you having us on
1:16:07
really appreciate you having us on
1:16:07
really appreciate you having us on oh I take it take care

Current & Future State of Voice 🔊 Technology || C# Corner

CSharpCorner

How Agents Can Simplify Your Vacation Requests

How AI Agents Enhance Employee Experience and Productivity

The Future of AI Engineering: Low Code, No Code Revolution

Unlocking AI: Transforming Everyday Tasks with Agents

शिक्षक दिवस विशेष – गुरु वंदन @CSharpCorner

Breaking Down Silos: The Key to Successful Cloud Projects

The Importance of Cybersecurity Assessments

Why Compliance is Crucial for Cybersecurity

Understanding Zero Trust: The Key to Cybersecurity

The Future of Cybersecurity in an AI World

AI in Retail: Developer Tools, APIs & MLOps Strategies

Citus Con: An Event for Postgres 2023 – Americas Livestream

Code Quality Conference 2023

Software Architecture Conference 2023

Why Is Nobody Talking About Good Low Cap Coins

Up next in 10

Current & Future State of Voice 🔊 Technology || C# Corner

CSharpCorner