API TTS - Text To Speech

• URL

  https://viettelgroup.ai/voice/api/tts/v1/rest/syn

• Method

  http POST

• Header Parameters

  1. header "Content-Type: application/json" 2. header "token: token_id"

  Để có "token id" bạn có thể đăng ký tài khoản tại viettelgroup.ai, sau đó login, rồi vào menu token để tạo

Định dạng data gửi lên: Json (default)

{
    "text": "hệ thống tổng hợp tiếng nói trung tâm không gian mạng",
    "voice": "doanngocle",
    "id": "2",
    "without_filter": false,
    "speed": 1.0,
    "tts_return_option": 2
}

Mô tả các tham số:

Success Response:

• Code: 200
  
• Content: inputstream (audio wav format)

Giọng đọc hiện có

Bạn có thể lấy danh sách giọng đọc mới nhất thông qua API: https://viettelgroup.ai/voice/api/tts/v1/rest/voices

example curl:

tạo file data.json như sau:

{
    "text": "Hệ thống tổng hợp tiếng nói trung tâm không gian mạng",
    "voice": "doanngocle",
    "id": "2",
    "without_filter": false,
    "speed": 1.0,
    "tts_return_option": 2
}

command:

curl --header "Content-Type: application/json" --header "token: demo_token"   --request POST   --data "@data.json" https://viettelgroup.ai/voice/api/tts/v1/rest/syn > example.wav
#if encounter SSL error because of https certificate, add option '-k' to  make insecure connections   (this will expose your application to security risks, such as man-in-the-middle attacks.)

output trả về là file: example.wav

example python

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# Import the json library
import json
import requests
import urllib

url = "https://viettelgroup.ai/voice/api/tts/v1/rest/syn"
data = {"text": "hôm nay là thứ mấy", "voice": "doanngocle", "id": "2", "without_filter": False, "speed": 1.0, "tts_return_option": 2}
headers = {'Content-type': 'application/json', 'token': 'demo_token'}
response = requests.post(url, data=json.dumps(data), headers=headers)
#if encounter SSL error because of https certificate, please comment out above line and use the line below to  make insecure connections   (this will expose your application to security risks, such as man-in-the-middle attacks.)
#response = requests.post(url, data=json.dumps(data), headers=headers, verify=False)
# Headers is a dictionary
print(response.headers)
# Get status_code.
print(response.status_code)
# Get the response data as a python object.
data = response.content
f = open("/home/tanpk/log/tanpk.wav", "wb")
f.write(data)
f.close()

example java:

public class TestAPI {
    public static InputStream getTTS(String apiUrl, String datajson) throws IOException {
        HttpClient httpClient = HttpClientBuilder.create().build();
        HttpPost request = new HttpPost(apiUrl);
        StringEntity body = new StringEntity(datajson, "UTF-8");
        /*add content-type, token into header request */
        request.addHeader("content-type", "application/json;charset=UTF-8");
        request.addHeader("token", "demo_token");
        request.getRequestLine();
        request.setEntity(body);
        HttpResponse response = httpClient.execute(request);
        System.err.println(response);
        return response.getEntity().getContent();
    }
    public static void main(String[] args) throws IOException {
        /*
        * create datajon format
        * text need to endcode by URLencode
        * */
        String datajson = "{\"text\":\"xin chào các bạn"," +
                "\"voice\":\"doanngocle\"," +
                "\"id\":\"3\"," +
                "\"without_filter\":false," +
                "\"speed\":1.0," +
                "\"tts_return_option\":2}";
        InputStream result = getTTS("https://viettelgroup.ai/voice/api/tts/v1/rest/syn", datajson);
        /* you can write result to file to use */
    }
}