Audio Need Text-to-speech help

Hello.

I have been using unity as a hobby for a few years and have a few completed apps under my belt, but still a novice really. I need some advice on text to speech software, I hope this is the right forum for it.

I have a disabled son who is 8 years but non verbal. He uses software at school where if he touches the picture on an iPad the corresponding word is read out ie; he touches the cup picture and he hears the word "juice". I wanted to make a similar app for his android tablets at home, so he has a voice.
The coding and layout I'm fine with, but I thought I could use an artificial voice for the words read out. It should be easier than trying to record my own voice for it lol.
Does anyone have any recommendations?
I have seen Amazon Polly but not sure if I can save the audio and use it as a clip in my app.

Thanks.

 

Apparently you can use google translate tts to do this using WWW function. I have not yet tested this, but here is a link to a youtube video I ran across.

I will test this soon, and if I come across something else I will let you know!

 

That's great, thank you for the reply.

 

I have a working version. The URL link provided in the video was caught by google and triggered a captcha.
use this url instead and it works great!
"https://translate.google.com/translate_tts?ie=UTF-8&client=tw-ob&tl=en&q="
This is what my TTS function looks like:

Code (CSharp):

1. public IEnumerator GetTTS()
2.         {
3.            
4.             Debug.Log(TSSwords);
5.              // Remove the "spaces" in excess
6.                  Regex rgx = new Regex ("\\s+");
7.                  // Replace the "spaces" with "% 20" for the link Can be interpreted
8.                  result = rgx.Replace (ITCwords, "%20");
9.                   Debug.Log(result);                            
10.                  url = "https://translate.google.com/translate_tts?ie=UTF-8&client=tw-ob&tl=en&q=" + result;
11.                  WWW www = new WWW (url);
12.                  yield return www;
13.                  GetComponent<AudioSource>().clip = www.GetAudioClip (false, true, AudioType.MPEG);
14.                  GetComponent<AudioSource>().Play ();
15.         }

 

If you are looking for a easy Text to audio converter (with over 400+ human sounding voices & languages) you can give texttoaudiospeech.com a try. I use it for my own projects.

 

Personally, for me, Audext is the best service to convert audio files into text. It has 2 options: manual or automatic transcription but both are pretty accurate in use.

 

In my opinion, the best text-to-speech converter for Indian languages and the cheapest is Spylift. Know more about it here or access it directly here

 

I feel like the best text to speech AI for Indian languages is Wavel as it has multiple accents and voices specifically designated for native language conversions and gives off a more authentic and natural tone

 

Hello!

Firstly, it's heartwarming to see the effort you're putting in to make things easier for your son. Regarding your query, while Amazon Polly is a popular choice, you might want to check out Rythmex. It's a transcription tool, but many transcription services often have complementary text-to-speech features. Not only can you use it for converting text to speech, but it might also offer high-quality voices that sound more natural than some other services. It should allow you to save the audio and use it as a clip in your app.

Best of luck with your project, and I hope it brings great value to your son's experience!

 

Speaktor is also very useful for text to speech translation. I also use Transkriptor for audio to text translation.
You can also use Transkriptor's API in the codes you develop for Android. There is a developer page for that too: https://developer.transkriptor.com/docs

 

- For a good tts service integrated already (with scene examples in their libraries) in unity and for free I would use wit from meta. https://wit.ai/ If my memory is good, voices can be generated offline in english but for other lanuages has to be ondemand and you need to register (free).

- If you want a more control ot realistic voices you have to go for a paid service. Amazon Polly, Google https://cloud.google.com/text-to-speech , ... they have a wide variaety in voices languages and settings. You have to signup on those serviecs, create a credentials an integrate those services (can be done with UnityWebRequest).

- Recently, Im testing the brandnew TTS service from OpenAI and it sounds very good quality for a similar price.