0:00
So recently OpenAI, which is the parent company of ChatGPT, has announced some
0:06
exciting features related to ChatGPT and other AI tools. One of the news is that they have opened the API key for
0:14
their text-to-speech services. They already had a transcription AI called WhisperAI, which was a speech-to-text AI tool
0:23
Now this one is text-to-speech, which means you can put in the text and it
0:29
will talk back to you. It will convert that text into audio and the audio output is human-like
0:35
So this text-to-speech service, which offers human-like voices, has got a lot of use cases
0:41
One of the use cases which I can see in my field of work in content creation is I can
0:47
easily use it for voice-overs and tutorial videos where I don't feel like recording my
0:52
own audio because maybe I'm traveling or I don't have my audio setup with me
0:56
And there are other use cases as well. Bloggers can also use for converting their articles into audio formats and distribute
1:04
them as podcasts and maybe companies can also use it for IVR, automated call systems
1:11
The possibilities are endless. But for people like you and me who will require a user interface to use something
1:18
like this, there is a tool from a developer who created Mac Whisper and MacGPT, which
1:26
is available for free to download. And these are awesome tools I have been using it
1:31
This new tool called Voices now lets you use the OpenAI's API key for text-to-speech
1:39
services. Now there are many tools available in the market right now and the best one is 11 Labs
1:45
It has realistic human voices. Now it has got a competition from OpenAI in the form of this tool
1:53
Let me turn on the screen recording. So here we go. This is Voices application which is currently available for Mac only and it is from the
2:01
developer who has created Whisper tool, Mac Whisper, which is an amazing transcription
2:07
tool which uses OpenAI's API key, the Whisper AI's API key to convert your voice into
2:14
text, basically transcribe and gives you high quality text. Now the same developer has created this Voices which is high quality text-to-speech
2:22
using OpenAI. You can download it for free. Once you download it, you will get an email with that download link and when you
2:31
download it like I have, the file comes in zip format. You can extract it, then you can install it
2:37
It will ask you your OpenAI API key which you can get from the OpenAI's website
2:45
So once you are on this page, you can create your API key and before that, make sure that
2:51
you have signed up and you have got free credits. If you haven't got free credits, you will have to purchase one
2:56
You can easily do so by going to the Billings area where you can add your payment
3:04
method and then you can buy credits. The minimum you can get is $5 and it's between $5 to $100
3:10
So I added $10 to my account and it lasts quite a long time
3:15
I use it a lot and you can see that the invoice is just one cent
3:21
Now we will go to API keys and I will click on create new
3:25
I will give it a description so that I will be able to recognize it
3:31
Once I tap on create API key, I'll copy it from here and then I'll click on done
3:39
And once I'm done, I will not be able to copy it again
3:44
So make sure you copy that from that screen before clicking on done button because you
3:51
will not be able to copy it again and do not share it with anyone else
3:55
So we have got our API key. I'll come back to Voices app which I just downloaded and installed and in the API key
4:05
section, I will paste my API key. Okay, so this is the interface of this application and on the right hand side, I can see it's
4:17
Oli. There's a drop down where I can select the different voices and then there is quality
4:26
of the output and the output format. And this is the place where I can type anything or I can put my text which will be
4:36
inverted. So now I click on create audio
4:47
This is a text to see how this thing works. Is it human enough? Okay, that sounds good
4:57
I'll choose another voice. This is a text to see how this thing
5:03
I'll have to create again. This is a text to see how this thing works
5:10
Is it human enough? The voices are quite high quality. Let's select TTS1HD
5:20
So I'm assuming HD means high definition. What else? This is a text to see how this thing works
5:27
Is it human enough? So it understands where the question mark is. So it's a question
5:32
So you will have to say it that way. Let me remove the question mark and then create again
5:43
This is a text to see how this thing works. Is it human enough? So there's a change, slight change because there is no question mark
5:50
So it's quite smart. Let's check other voices as well. As of now, there are only six voices
5:59
Maybe in future they will add some more. This is a text to see how this thing works
6:04
Is it human enough? This is a text to see how this thing works. Is it human enough
6:09
Nova. And now we are on the sixth voice. This is a text to see how this thing works
6:18
Is it human enough? OK, let me copy something from a website
6:24
Let's see how this sounds. I'll go with Echo. Let's go with Fable
6:30
The dollar's rebound extended for a third day on Wednesday after some Federal Reserve policymakers left the door open to further rate hikes
6:38
as traders looked to a speech from Chair Jerome Powell on the central bank's future policy path
6:45
The greenback, which hit a seven week low at the start of the week
6:49
in the wake of the Fed's decision to hold its policy rate steady and on data pointing to a cooling US labor market
6:57
has found a flaw as markets remain at odds over whether a peak in US rates has been reached
7:02
and how soon the Fed could begin easing monetary conditions. OK, so I just noticed that on the interface itself
7:09
it shows you how much money is being charged and it's quite nominal
7:16
So that's it. I think it's pretty cool. It's pretty high quality and not so expensive
7:21
All the other tools available in the market right now are on monthly subscription basis
7:27
There's no plan, at least no high quality human like voices service
7:32
that offers pay as you go model. The best, the next best after this one is 11 Labs
7:40
but it is also on a subscription basis. But thanks to this application by this developer
7:46
who also created Mac, GPT and Mac Whisper. So thanks for creating this tool
7:53
so that people like me who are into productivity and content creation can use a tool which is pay as you go
7:59
So as you can see, it's not much you will have to pay. So that's about it
8:04
All the information about this tool, the tutorial, the written article link is in the description of this video
8:11
You can check it out and follow for more videos like this