Text To Speech API

All the available APIs needed to generate audio from text.

Convai's Text-To-Speech API endpoint

POST https://api.convai.com/tts/

This endpoint is called to convert a user provided text to audio in the voice mentioned by the user.

Headers

Request Body

403: Forbidden Error with the user's Convai API Key.

{
    "ERROR" : "api_key error."
}

200: OK transcript not specified in the request body.

{
    "result" : "No transcript found in the request"
}

200: OK voice not specified in the request body.

{
    "result" : "No voice found in the request"
}

403: Forbidden Incorrect Convai API Key provided by the user.

{
    "API_ERROR" : "Invalid API key provided."
}

400: Bad Request The mentioned voice doesn't support the requested audio encoding format. Please refer to the voice-encoding table below.

{
    "result" : "Incorrect encoding for the mentioned voice."
}

500: Internal Server Error Error encountered while processing the TTS request.

{
    "result" : "Error encountered while processing the TTS request."
}

200: OK successful response : an audio file

List of available public voices and their supported languages:

All the voices support the wav format. Only GCP Voices support both wav and mp3 formats.

Please provide the Voice Code in the voice field of the payload.

Please note if an unsupported audio encoding is specified along with a voice in the request body, it will result in a 400 error response.

The above section only lists the public voices. As Convai supports third-party integrations, to get the full list of voices available to a user, please check out Voice List API.

List of available voices:

Here are some sample codes to demonstrate the request format for the endpoint -->

Python

import requests
import json

url = "https://api.convai.com/tts/"

payload = json.dumps({
  "transcript": "Hello World from us",
  "voice": "WUMale 1",
  "filename": "testAudio",
  "encoding": "mp3"
})
headers = {
  'CONVAI-API-KEY': '<your api key>',
  'Content-Type': 'application/json'
}

response = requests.request("POST", url, headers=headers, data=payload)

with open('testAudio.mp3','wb') as f:
  f.write(response.content)

cURL

curl --location --request POST 'https://api.convai.com/tts' \
--header 'CONVAI-API-KEY: <your api key>' \
--header 'Content-Type: application/json' \
--data-raw '{  
    "transcript": "Hello World from us",
    "voice": "WUMale 1",
    "filename": "testAudio",
    "encoding" : "mp3"
}' \
--output testAudio.mp3