GLM TTS API 接口、参数 & 代码示例

GLM-TTS 是一款“超拟人”的语音合成大模型。它突破了传统的语音合成框架，让合成的语音具备真实的情感和生命力。

模型概览与技术架构：

• 核心定位：以新一代智谱语音大模型为核心的语音合成系统。它能够根据上下文智能地预判文本的情绪与语调，显著提升语音的自然度与表现力。
• 架构设计：采用两阶段生成架构：
  • 第一阶段由基于大语言模型（LLM）的架构将输入文本转换为语音 Token 序列。
  • 第二阶段利用 Flow Matching（流匹配）模型将 Token 序列转化为高质量的音频波形（或梅尔谱）。
• 强化学习优化：在训练中首次引入了基于 GRPO（Group Relative Policy Optimization）的多奖励强化学习方案，在公开评测的“字错误率（CER）”和“情感表达”上取得了开源领域的 SOTA（业内顶尖）表现。

核心功能与特性：

• 超拟人与情感表达增强：模型具备极强的上下文理解能力，不再是机械式地朗读，而是能够实现“角色化演绎”以及根据内容动态调整情感。
• 低延迟流式接口（Streaming）：
• 支持非流式和流式两种接口。
• 流式接口返回响应极快，首帧响应速度可达到 400ms 以内，非常适合实时、低延迟的交互式场景。
• 动态参数调节：调用时支持随心调节语速（speed）和音量（volume）等参数，以满足复杂的业务场景要求。
• 多音色与音色复刻（Clone）：
• 平台提供了如 tongtong（彤彤）、chuichui（锤锤）、xiaochen（小陈）等多种系统默认音色，还包括动动动物圈系列（如 jam, kazi, douji, luodo 等）。
• 配合 glm-tts-clone 模型，它还支持高效的“音色复刻”能力。

主要推荐应用场景：

• 智能客服：依托超拟人语音的情感适配，提供全链路的柔性服务，降低用户的抵触感。
• 有声阅读：通过“角色化演绎+情感随内容动态调整”的能力，打造沉浸式的个性化听书体验。
• 智能交互助手：为智能硬件或车载助理赋予真实情感与场景化语调，使其摆脱冰冷的“工具属性”。
• 教育领域与职场办公：用于场景化教学（提升学习沉浸感）以及会议纪要转语音、邮件/文档播报等场景。
• 文旅领域：可替代传统导游，应用于景区智能导览、酒店智能服务等。

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.wav 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.wav \
--data '{
    "model": "z-ai/glm-tts",
    "input": "你好，今天天气怎么样。",
    "voice": "tongtong",
    "response_format": "wav"
}'

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": "z-ai/glm-tts",
  "input": "你好，今天天气怎么样。",
  "voice": "tongtong",
  "response_format": "wav"
] 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.wav"))
  }
})

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": "z-ai/glm-tts",
    "input": "你好，今天天气怎么样。",
    "voice": "tongtong",
    "response_format": "wav"
  });
  request.headers.addAll(headers);

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

  if (response.statusCode == 200) {
    var bytes = await response.stream.toBytes();
    File('speech.wav').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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}"

response = http.request(request)
File.write("speech.wav", 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": "z-ai/glm-tts",
    "input": "你好，今天天气怎么样。",
    "voice": "tongtong",
    "response_format": "wav"
  });

  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.wav", 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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}");

FILE *fp = fopen("speech.wav", "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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}")

  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.wav", 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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}", null, "application/json");

var response = await client.SendAsync(request);

response.EnsureSuccessStatusCode();

await File.WriteAllBytesAsync("speech.wav", 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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}", ParameterType.RequestBody);

var response = client.Execute(request);

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

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

let data = JSON.stringify({
  "model": "z-ai/glm-tts",
  "input": "你好，今天天气怎么样。",
  "voice": "tongtong",
  "response_format": "wav"
});

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.wav', 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\":\"z-ai/glm-tts\",\"input\":\"你好，今天天气怎么样。\",\"voice\":\"tongtong\",\"response_format\":\"wav\"}");

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.wav"), 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": "z-ai/glm-tts",
  "input": "你好，今天天气怎么样。",
  "voice": "tongtong",
  "response_format": "wav"
}';

$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.wav', $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' => 'z-ai/glm-tts',
    'input' => '你好，今天天气怎么样。',
    'voice' => 'tongtong',
    'response_format' => 'wav',
  ]),
  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.wav', $response);
}

import requests

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

payload = {
  "model": "z-ai/glm-tts",
  "input": "你好，今天天气怎么样。",
  "voice": "tongtong",
  "response_format": "wav"
}

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.wav", "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="z-ai/glm-tts",
  input="你好，今天天气怎么样。",
  voice="tongtong",
  response_format="wav",
)

response.write_to_file("speech.wav")