Generate TTS Audio (Text-to-Speech)

Global Server

POST

/api/v1/video/send_tts

If you want to use our TTS engine to generate an audio file and use it later for the "create" method for video synthesis, use this API.
You do not neccessarily need this API if you choose to upload your own audio to drive the lip motion of avatars. Note there are two ways to provide the audio for avatar video generation.

upload your own audio
Use the TTS that we provide (This API endpoing).

You can choose from public available TTS engines (currently we provdie Microsoft Azure and Elevenlabs), or use your custom trained TTS (obtained by the methods listed in user_voice section).

Request

Authorization

Provide your bearer token in the

Authorization

header when making requests to protected resources.

Example:

Authorization: Bearer ********************

Body Params application/json

msg

string

required

the text for TTS

tts_id

string

optional

Use this if you want to use public TTS. The id of the voice. This is obtained from the the return of /api/v1/anchor/voice_list. Use data -> children -> value. Example: 66dc61ec5148817d26f5b79e

user_voice_id

string

optional

Use this if you want to use your cloned voice. The id of the voice clone training result. (obtained from the api of /api/v1/userVoice/completedRecord)
You can provide either tts_id or user_voice_id.
You must provide "country" and "region" if you set this parameter.

speechRate

number

required

1 is normal speed. The larger this value, the faster the speech.

country

string

optional

The code of the country.
When you use "user_voice_id", you must provide this.
Valid options include: zh, en, ar, ja, es, de, fr, etc. For all available choices, refere to the results returned by /v1/anchor/voice_list method

region

string

optional

The code of the language.
Valid options include: CN, US, AE etc. So the system can combine en-US (US accent English), zh-CN (Mainland accent Chinese), ar-AE (UAE accent Arabic) etc.
For all available choices, refere to the results returned by /v1/anchor/voice_list method. When you provide "country", you must provide this.

Example

{
  "msg": "hello every one，i am eric，nice to see you！",
  "tts_id": "66503e3d9d7679eb9b640343",
  "speechRate": 1.1,
  "country": "en",
  "region": "US"
}

Request samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://video.a2e.ai/api/v1/video/send_tts' \
--header 'Content-Type: application/json' \
--data-raw '{"msg":"hello every one，i am eric，nice to see you！",
 "tts_id":"66503e3d9d7679eb9b640343",
 "speechRate":1.1,
 "country": "en",
 "region": "US"}'

Responses

🟢200send_tts

application/json

Body

code

integer

required

data

string

required

The url to download the audio file

Example

{
  "code": 0,
  "data": "https://dh24as48lv9ce.cloudfront.net/ai/speech/tts/2023-10-11/en-US-JennyMultilingualNeural/tts_f44zzb8h.wav"
}

Modified at 2025-03-02 20:10:26

List Ongoing Voice Clone Tasks

Get Details of a Voice