Vera Voice API (Beta)

Here is API documentation for TTS service from VeraVoice. First of all, you need to pass registration. It's not open for everybody, so you need to write an email to apirequest@veravoice.ai and describe your use case.

After you receive username and password, you need to get auth-token:

import requests


login_url = "https://api.veravoice.ai/api-token-auth/"
req = requests.post(login_url, data={"username": "you_username", "password": "your_password"})
token = req.json()["token"]

Then you can call TTS API like this:

data = {
  "speaker": "archer", 
  "text": "Your Text"
 }
req = requests.post("https://api.veravoice.ai/api/tts", 
                    data=data, headers={"Authorization": f"Token {token}"})

API endpoint: https://api.veravoice.ai/api/tts Method: POST

Parameters

• speaker - Speaker, you can get a list of all available speakers at the link: https://api.veravoice.ai/api/speakers/
• text - The text you want to synthesize, e. g. "hello". 1500 characters limitation. If you're going to add stress to the word, use "+" before the stressed vowel, e. g. "hell+o". The numbers need to be written as words, e. g. “10” - “ten”.
• frame_rate - integer, valid values are from 0 to 1000, if you pass this parameter, it returns the array of samples of audio (passed value of samples per second) and audio in base64 in JSON-response. You can get it like this: req.json()["array_of_samples"], and get audio bytes like this:

import base64
audio_bytes = base64.b64decode(req.json()["audio"])

Your requests' statistics available at this https://api.veravoice.ai/api/requests/.