CosyVoice V3 Plus API 接口、参数 & 代码示例

CosyVoice V3 Plus 是通义实验室推出的最新一代生成式语音合成（TTS）大模型。该模型代表了目前 CosyVoice 系列中合成效果最佳、综合能力最强的旗舰版本。

核心功能与亮点

• 零样本声音复刻（Zero-Shot Voice Cloning）： 仅需提供 10~20秒 的发音人样本音频，即可在不进行任何额外训练的情况下，快速生成高度相似、自然且极具表现力的定制音色。
• 声音设计（Voice Design）： 支持通过纯文本描述（如描述性别、年龄、音色特征、多语言等维度）来无中生有地创造出全新的、符合特定人设的虚拟音色。
• 方言与多语言支持： 相比前代，v3-plus 大幅增强了多语言及中国各地方言的复刻与操控能力。

语言与方言支持能力

音频输出规格

CosyVoice V3 Plus 提供了极高的音频兼容性，能够满足从电话客服到超高清广播等各种场景的需求：

• 输出格式： pcm、wav、mp3、opus
• 采样率（Sample Rate）： 支持广泛的采样率调节，包括 8kHz、16kHz、22.05kHz、24kHz、44.1kHz、48kHz。

适用场景与方案选择

CosyVoice V3 Plus 主要服务于以下场景：

1. 标准语音合成： 当内置的官方音色库（包含多语种、各年龄段、客服/小说等丰富人设）已经能满足需求时，可直接调用进行高品质文本转语音。
2. 个性化复刻/设计： 用于需要打造品牌专属声音、复刻特定人物、或为游戏/动漫角色创建全新音色。

💡 CosyVoice V3 Plus 目前仅支持 longanyang（龙安洋，阳光大男孩） 和 longanhuan（龙安欢，欢脱元气女） 2 个系统音色 ID。

此 API 接口兼容 OpenAI 的 Text-to-Speech 接口规范，可直接使用 OpenAI 的 SDK 来调用。仅需替换以下配置即可：

1. base_url 替换为 https://wcode.net/api/gpt/v1
2. api_key 替换为从 https://platform.wcode.net 获取到的 API Key

具体可参考下方的各编程语言代码示例中的 OpenAI SDK 调用示例。

# TODO: 以下代码中的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
# 响应为音频二进制，可使用 --output speech.mp3 保存文件

curl --request POST 'https://wcode.net/api/gpt/v1/audio/speech' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer API_KEY' \
--output speech.mp3 \
--data '{
    "model": "tongyi/cosyvoice-v3-plus",
    "input": "你好，今天天气怎么样。",
    "voice": "longanyang"
}'

import Foundation

let headers = [
  "Authorization": "Bearer API_KEY",  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  "content-type": "application/json"
]
let parameters = [
  "model": "tongyi/cosyvoice-v3-plus",
  "input": "你好，今天天气怎么样。",
  "voice": "longanyang"
] as [String : Any]

let postData = JSONSerialization.data(withJSONObject: parameters, options: [])

let request = NSMutableURLRequest(url: NSURL(string: "https://wcode.net/api/gpt/v1/audio/speech")! as URL,
                                        cachePolicy: .useProtocolCachePolicy,
                                    timeoutInterval: 60.0)
request.httpMethod = "POST"
request.allHTTPHeaderFields = headers
request.httpBody = postData as Data

let session = URLSession.shared
let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in
  if (error != nil) {
    print(error as Any)
  } else if let data = data {
    try? data.write(to: URL(fileURLWithPath: "speech.mp3"))
  }
})

dataTask.resume()

import 'dart:convert';
import 'dart:io';
import 'package:http/http.dart' as http;

Future<void> main() async {
  var headers = {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer API_KEY'  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  };
  var request = http.Request('POST', Uri.parse('https://wcode.net/api/gpt/v1/audio/speech'));
  request.body = json.encode({
    "model": "tongyi/cosyvoice-v3-plus",
    "input": "你好，今天天气怎么样。",
    "voice": "longanyang"
  });
  request.headers.addAll(headers);

  http.StreamedResponse response = await request.send();

  if (response.statusCode == 200) {
    var bytes = await response.stream.toBytes();
    File('speech.mp3').writeAsBytesSync(bytes);
  }
  else {
    print(response.reasonPhrase);
  }
}

require 'uri'
require 'net/http'

url = URI("https://wcode.net/api/gpt/v1/audio/speech")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer API_KEY'  # TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
request["content-type"] = 'application/json'
request.body = "{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}"

response = http.request(request)
File.write("speech.mp3", response.body) if response.is_a?(Net::HTTPSuccess)

use serde_json::json;
use reqwest;
use std::fs;

#[tokio::main]
pub async fn main() {
  let url = "https://wcode.net/api/gpt/v1/audio/speech";

  let payload = json!({
    "model": "tongyi/cosyvoice-v3-plus",
    "input": "你好，今天天气怎么样。",
    "voice": "longanyang"
  });

  let mut headers = reqwest::header::HeaderMap::new();
  headers.insert("Authorization", "Bearer API_KEY".parse().unwrap());  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  headers.insert("content-type", "application/json".parse().unwrap());

  let client = reqwest::Client::new();
  let response = client.post(url)
    .headers(headers)
    .json(&payload)
    .send()
    .await
    .unwrap();

  let audio_bytes = response.bytes().await.unwrap();
  fs::write("speech.mp3", audio_bytes).unwrap();
}

#include <stdio.h>
#include <curl/curl.h>

static size_t write_callback(void *contents, size_t size, size_t nmemb, void *userp) {
  return fwrite(contents, size, nmemb, (FILE *)userp);
}

CURL *hnd = curl_easy_init();

curl_easy_setopt(hnd, CURLOPT_CUSTOMREQUEST, "POST");
curl_easy_setopt(hnd, CURLOPT_URL, "https://wcode.net/api/gpt/v1/audio/speech");

struct curl_slist *headers = NULL;
headers = curl_slist_append(headers, "Authorization: Bearer API_KEY");  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
headers = curl_slist_append(headers, "content-type: application/json");
curl_easy_setopt(hnd, CURLOPT_HTTPHEADER, headers);

curl_easy_setopt(hnd, CURLOPT_POSTFIELDS, "{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}");

FILE *fp = fopen("speech.mp3", "wb");
curl_easy_setopt(hnd, CURLOPT_WRITEFUNCTION, write_callback);
curl_easy_setopt(hnd, CURLOPT_WRITEDATA, fp);

CURLcode ret = curl_easy_perform(hnd);

fclose(fp);
curl_easy_cleanup(hnd);
curl_slist_free_all(headers);

package main

import (
  "fmt"
  "os"
  "strings"
  "net/http"
  "io"
)

func main() {
  url := "https://wcode.net/api/gpt/v1/audio/speech"

  payload := strings.NewReader("{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}")

  req, _ := http.NewRequest("POST", url, payload)

  req.Header.Add("Authorization", "Bearer API_KEY")  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  req.Header.Add("content-type", "application/json")

  res, _ := http.DefaultClient.Do(req)

  defer res.Body.Close()
  body, _ := io.ReadAll(res.Body)

  os.WriteFile("speech.mp3", body, 0644)
  fmt.Println(res.Status)
}

using System.Net.Http.Headers;


var client = new HttpClient();

var request = new HttpRequestMessage(HttpMethod.Post, "https://wcode.net/api/gpt/v1/audio/speech");

request.Headers.Add("Authorization", "Bearer API_KEY");  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net

request.Content = new StringContent("{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}", null, "application/json");

var response = await client.SendAsync(request);

response.EnsureSuccessStatusCode();

await File.WriteAllBytesAsync("speech.mp3", await response.Content.ReadAsByteArrayAsync());

var client = new RestClient("https://wcode.net/api/gpt/v1/audio/speech");

var request = new RestRequest("", Method.Post);

request.AddHeader("Authorization", "Bearer API_KEY");  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net

request.AddHeader("content-type", "application/json");

request.AddParameter("application/json", "{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}", ParameterType.RequestBody);

var response = client.Execute(request);

if (response.IsSuccessful && response.RawBytes != null)
    await File.WriteAllBytesAsync("speech.mp3", response.RawBytes);

const axios = require('axios');
const fs = require('fs');

let data = JSON.stringify({
  "model": "tongyi/cosyvoice-v3-plus",
  "input": "你好，今天天气怎么样。",
  "voice": "longanyang"
});

let config = {
  method: 'post',
  maxBodyLength: Infinity,
  url: 'https://wcode.net/api/gpt/v1/audio/speech',
  responseType: 'arraybuffer',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer API_KEY'  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  },
  data : data
};

axios.request(config).then((response) => {
  fs.writeFileSync('speech.mp3', response.data);
}).catch((error) => {
  console.log(error);
});

import java.nio.file.Files;
import java.nio.file.Path;

OkHttpClient client = new OkHttpClient();

MediaType mediaType = MediaType.parse("application/json");

RequestBody body = RequestBody.create(mediaType, "{\"model\":\"tongyi/cosyvoice-v3-plus\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"longanyang\"}");

Request request = new Request.Builder()
  .url("https://wcode.net/api/gpt/v1/audio/speech")
  .post(body)
  .addHeader("Authorization", "Bearer API_KEY")  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  .addHeader("content-type", "application/json")
  .build();

try (Response response = client.newCall(request).execute()) {
  if (response.isSuccessful() && response.body() != null) {
    Files.write(Path.of("speech.mp3"), response.body().bytes());
  }
}

$client = new \GuzzleHttp\Client();

$headers = [
  'Content-Type' => 'application/json',
  'Authorization' => 'Bearer API_KEY',  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
];

$body = '{
  "model": "tongyi/cosyvoice-v3-plus",
  "input": "你好，今天天气怎么样。",
  "voice": "longanyang"
}';

$request = new \GuzzleHttp\Psr7\Request('POST', 'https://wcode.net/api/gpt/v1/audio/speech', $headers, $body);

$response = $client->sendAsync($request)->wait();

file_put_contents('speech.mp3', $response->getBody()->getContents());

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://wcode.net/api/gpt/v1/audio/speech",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 5,
  CURLOPT_TIMEOUT => 300,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'model' => 'tongyi/cosyvoice-v3-plus',
    'input' => '你好，今天天气怎么样。',
    'voice' => 'longanyang',
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer API_KEY",  // TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
    "content-type: application/json",
  ],
]);

$response = curl_exec($curl);
$error = curl_error($curl);

curl_close($curl);

if ($error) {
  echo "cURL Error #:" . $error;
} else {
  file_put_contents('speech.mp3', $response);
}

import requests

url = "https://wcode.net/api/gpt/v1/audio/speech"

payload = {
  "model": "tongyi/cosyvoice-v3-plus",
  "input": "你好，今天天气怎么样。",
  "voice": "longanyang"
}

headers = {
  "Authorization": "Bearer API_KEY",  # TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
  "content-type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

with open("speech.mp3", "wb") as f:
  f.write(response.content)

from openai import OpenAI

client = OpenAI(
  base_url="https://wcode.net/api/gpt/v1",
  api_key="API_KEY"  # TODO: 这里的 API_KEY 需要替换，获取 API Key 入口：https://platform.wcode.net
)

response = client.audio.speech.create(
  model="tongyi/cosyvoice-v3-plus",
  input="你好，今天天气怎么样。",
  voice="longanyang",
)

response.write_to_file("speech.mp3")