@amaster.ai/tts-client 1.1.0-beta.7 → 1.1.0-beta.71

package/README.md

@@ -1,126 +1,270 @@
-# TTS Realtime WebSocket Client
+# TTS Realtime WebSocket Client SDK
 
-一个基于 WebSocket 的实时文本转语音（TTS）客户端 SDK，用于对接 `qwen-tts` 实时语音合成服务，支持流式音频返回与自动播放。
+基于 WebSocket 的实时文本转语音（TTS）客户端 SDK，用于对接 `qwen-tts` 实时语音合成服务，支持流式音频返回与自动播放。
+
+---
 
 ## 特性
 
-* WebSocket 实时 TTS
-* 支持多种音频格式（`pcm` / `mp3` / `wav` / `opus`）
-* 支持自动播放或手动播放
-* 支持音频流分片回调
-* 浏览器原生 `AudioContext` 播放（PCM）
+- 🔊 WebSocket 实时 TTS
+- 🎵 支持多种音频格式（`pcm` / `mp3` / `wav` / `opus`）
+- ▶️ 支持自动播放或手动播放
+- 📦 支持音频流分片回调
+- 🎧 浏览器原生 `AudioContext` 播放（PCM）
+- 🎭 多种音色可选
+
+---
 
 ## 安装
 
-该 SDK 为纯前端实现，无额外依赖，直接拷贝或通过源码方式引入即可。
+```bash
+# npm
+npm install @amaster.ai/tts-client
 
-```ts
-import { createTTSClient } from "./tts-client";
+# pnpm
+pnpm add @amaster.ai/tts-client
+
+# yarn
+yarn add @amaster.ai/tts-client
 ```
 
+---
+
 ## 快速开始
 
+### 基础用法
+
 ```ts
+import { createTTSClient } from "@amaster.ai/tts-client";
+
 const tts = createTTSClient({
   voice: "Cherry",
+  autoPlay: true,
+  audioFormat: "pcm",
+  sampleRate: 24000,
   onReady() {
-    console.log("TTS ready");
+    console.log("TTS 已就绪");
   },
   onAudioStart() {
-    console.log("Audio start");
+    console.log("开始播放");
   },
   onAudioEnd() {
-    console.log("Audio end");
+    console.log("播放结束");
+  },
+  onAudioChunk(chunks) {
+    console.log("收到音频片段:", chunks.length);
   },
   onError(err) {
-    console.error(err);
+    console.error("TTS 错误:", err);
   },
 });
 
+// 建立连接
 await tts.connect();
-await tts.speak("你好，这是一个实时语音合成示例。");
-```
-
-## API 说明
 
-### createTTSClient(config)
+// 合成并播放语音
+await tts.speak("你好，欢迎使用实时语音合成服务。");
 
-创建一个 TTS 客户端实例。
+// 关闭连接
+// tts.close();
+```
 
-#### TTSClientConfig
+---
 
-| 参数           | 类型                          | 默认值                                                       | 说明             |
-| ------------ | --------------------------- | --------------------------------------------------------- | -------------- |
-| voice        | string                      | `Cherry`                                                  | 发音人名称          |
-| autoPlay     | boolean                     | `true`                                                    | 是否在音频接收完成后自动播放 |
-| audioFormat  | `pcm \| mp3 \| wav \| opus` | `pcm`                                                     | 音频格式           |
-| sampleRate   | number                      | `24000`                                                   | 采样率            |
-| onReady      | () => void                  | -                                                         | 会话初始化完成        |
-| onAudioStart | () => void                  | -                                                         | 音频开始播放         |
-| onAudioEnd   | () => void                  | -                                                         | 音频播放结束         |
-| onAudioChunk | (chunks: string[]) => void  | -                                                         | 接收到音频分片        |
-| onError      | (error: Error) => void      | -                                                         | 错误回调           |
+### React 完整示例
+
+```tsx
+import { useRef, useState } from "react";
+import { createTTSClient, type TTSClient } from "@amaster.ai/tts-client";
+
+const VoiceTypes = {
+  Cherry: "Cherry - 甜美女声",
+  Serena: "苏瑶 - 温柔小姐姐",
+  Ethan: "晨煦 - 标准普通话",
+  Chelsie: "千雪 - 二次元虚拟女友",
+  Peter: "天津话",
+};
+
+function TTSPlayer() {
+  const [voice, setVoice] = useState("Cherry");
+  const [connected, setConnected] = useState(false);
+  const [status, setStatus] = useState("disconnected");
+  const [text, setText] = useState("你好，欢迎使用通义千问实时语音合成服务。");
+  const clientRef = useRef<TTSClient | null>(null);
+
+  const connectTTS = () => {
+    if (clientRef.current) return;
+
+    const ttsClient = createTTSClient({
+      voice,
+      autoPlay: true,
+      audioFormat: "pcm",
+      sampleRate: 24000,
+      onReady: () => {
+        setConnected(true);
+        setStatus("connected");
+      },
+      onAudioStart: () => setStatus("playing"),
+      onAudioEnd: () => setStatus("connected"),
+      onAudioChunk: (chunks) => {
+        console.log("音频片段数:", chunks.length);
+      },
+      onError: (err) => {
+        console.error("TTS Error:", err);
+        setStatus("error");
+        setConnected(false);
+      },
+    });
+
+    ttsClient.connect();
+    clientRef.current = ttsClient;
+  };
+
+  const sendTTS = () => {
+    if (!text || !clientRef.current) return;
+    clientRef.current.speak(text);
+  };
+
+  const disconnectTTS = () => {
+    clientRef.current?.close();
+    clientRef.current = null;
+    setConnected(false);
+    setStatus("disconnected");
+  };
+
+  return (
+    <div>
+      <h3>🔊 实时语音合成（TTS）</h3>
+
+      <div>状态: {status}</div>
+
+      <div>
+        <label>音色:</label>
+        <select value={voice} onChange={(e) => setVoice(e.target.value)}>
+          {Object.entries(VoiceTypes).map(([key, label]) => (
+            <option key={key} value={key}>
+              {label}
+            </option>
+          ))}
+        </select>
+      </div>
+
+      <div>
+        <label>合成文本:</label>
+        <textarea rows={4} value={text} onChange={(e) => setText(e.target.value)} />
+      </div>
+
+      <div>
+        <button onClick={connectTTS} disabled={connected}>
+          1. 连接
+        </button>
+        <button onClick={sendTTS} disabled={!connected}>
+          2. 合成语音
+        </button>
+        <button onClick={disconnectTTS} disabled={!connected}>
+          断开
+        </button>
+      </div>
+    </div>
+  );
+}
+```
 
 ---
 
-### TTSClient
+## API 说明
 
-#### connect(): Promise<void>
+### `createTTSClient(config)`
 
-建立 WebSocket 连接并初始化会话。
+创建一个 TTS 客户端实例。
 
-```ts
-await tts.connect();
-```
+#### `TTSClientConfig`
 
-#### speak(text: string): Promise<void>
+| 参数             | 类型                                | 默认值     | 说明                                                                   |
+| ---------------- | ----------------------------------- | ---------- | ---------------------------------------------------------------------- |
+| `voice`          | `string`                            | `"Cherry"` | 发音人名称，可选值：`Cherry`, `Serena`, `Ethan`, `Chelsie`, `Peter` 等 |
+| `autoPlay`       | `boolean`                           | `true`     | 是否在音频接收完成后自动播放                                           |
+| `audioFormat`    | `"pcm" \| "mp3" \| "wav" \| "opus"` | `"pcm"`    | 音频格式，**注意：内置播放仅支持 `pcm`**                               |
+| `sampleRate`     | `number`                            | `24000`    | 采样率                                                                 |
+| `getAccessToken` | `() => string \| null`              | -          | 获取访问令牌（用于 WebSocket 认证）                                    |
+| `onReady`        | `() => void`                        | -          | 会话初始化完成回调                                                     |
+| `onAudioStart`   | `() => void`                        | -          | 音频开始播放回调                                                       |
+| `onAudioEnd`     | `() => void`                        | -          | 音频播放结束回调                                                       |
+| `onAudioChunk`   | `(chunks: string[]) => void`        | -          | 接收到音频分片回调                                                     |
+| `onError`        | `(error: Error) => void`            | -          | 错误回调                                                               |
 
-发送文本进行语音合成。
+#### `TTSClient`
 
 ```ts
-await tts.speak("Hello world");
+interface TTSClient {
+  connect(): Promise<void>; // 建立 WebSocket 连接
+  speak(text: string): Promise<void>; // 发送文本进行语音合成
+  play(): void; // 手动播放（autoPlay=false 时使用）
+  close(): void; // 关闭连接并释放资源
+}
 ```
 
-> 每次调用会清空上一次的音频缓存。
+---
 
-#### play(): void
+## 配合统一客户端使用
 
-手动播放当前已接收的音频数据（仅在 `autoPlay = false` 时有用）。
+推荐与 `@amaster.ai/client` 统一客户端一起使用，自动处理认证：
 
-```ts
-tts.play();
-```
+```tsx
+import { createClient } from "@amaster.ai/client";
 
-#### close(): void
+const client = createClient({
+  baseURL: "https://api.amaster.ai",
+});
 
-关闭 WebSocket 连接并释放音频资源。
+const ttsClient = client.tts({
+  voice: "Cherry",
+  autoPlay: true,
+  onReady() {
+    console.log("TTS 已就绪");
+  },
+  onAudioStart() {
+    console.log("开始播放");
+  },
+  onAudioEnd() {
+    console.log("播放结束");
+  },
+});
 
-```ts
-tts.close();
+await ttsClient.connect();
+await ttsClient.speak("你好，欢迎使用 Amaster AI！");
 ```
 
+---
+
 ## 音频说明
 
-* 当前内置播放逻辑 **仅支持 `pcm` 格式**
-* `pcm` 数据为 **16-bit little-endian 单声道**
-* 播放基于 `AudioContext`，仅支持浏览器环境
-* 若使用 `mp3 / wav / opus`，需自行实现解码与播放逻辑
+- 当前内置播放逻辑 **仅支持 `pcm` 格式**
+- `pcm` 数据为 **16-bit little-endian 单声道**
+- 播放基于 `AudioContext`，仅支持浏览器环境
+- 若使用 `mp3 / wav / opus`，需自行实现解码与播放逻辑
+
+---
 
 ## 运行环境
 
-* 浏览器（Chrome / Edge / Safari）
-* 需支持：
+- 浏览器（Chrome / Edge / Safari）
+- 需支持：
+  - WebSocket
+  - AudioContext
+  - atob
 
-  * WebSocket
-  * AudioContext
-  * atob
+---
 
 ## 注意事项
 
-* WebSocket 必须在 `connect()` 成功后才能调用 `speak()`
-* 多次 `speak()` 会覆盖之前的音频缓存
-* 自动播放依赖浏览器的自动播放策略，部分场景可能需要用户交互触发
+- WebSocket 必须在 `connect()` 成功后才能调用 `speak()`
+- 多次 `speak()` 会覆盖之前的音频缓存
+- 自动播放依赖浏览器的自动播放策略，部分场景可能需要用户交互触发
+- 建议在组件卸载时调用 `close()` 释放资源
+
+---
 
 ## License
 
-MIT
+MIT

package/dist/index.cjs

@@ -20,8 +20,7 @@ var __toCommonJS = (mod) => __copyProps(__defProp({}, "__esModule", { value: tru
 // src/index.ts
 var index_exports = {};
 __export(index_exports, {
-  createTTSClient: () => createTTSClient,
-  createTtsClient: () => createTTSClient
+  createTTSClient: () => tts_client_default
 });
 module.exports = __toCommonJS(index_exports);
 
@@ -29,7 +28,7 @@ module.exports = __toCommonJS(index_exports);
 var TTS_PATH = "/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime";
 function createTTSClient(config) {
   const {
-    path = TTS_PATH,
+    getAccessToken,
     voice = "Cherry",
     autoPlay = true,
     audioFormat = "pcm",
@@ -43,9 +42,18 @@ function createTTSClient(config) {
   let ws = null;
   let audioChunks = [];
   let audioContext = null;
+  let audioSource = null;
   async function connect() {
     return new Promise((resolve, reject) => {
-      ws = new WebSocket(path);
+      let wsUrl = TTS_PATH;
+      if (getAccessToken) {
+        const token = getAccessToken();
+        if (token) {
+          const separator = wsUrl.includes("?") ? "&" : "?";
+          wsUrl = `${wsUrl}${separator}token=${encodeURIComponent(token)}`;
+        }
+      }
+      ws = new WebSocket(wsUrl);
       ws.onopen = () => {
       };
       ws.onmessage = (event) => {
@@ -153,30 +161,43 @@ function createTTSClient(config) {
       source.connect(audioContext.destination);
       source.onended = () => onAudioEnd?.();
       source.start(0);
+      audioSource = source;
     } catch (err) {
       onError?.(err);
     }
   }
-  function close() {
-    if (ws) {
-      ws.close();
-      ws = null;
+  function stopAudio() {
+    if (audioSource) {
+      audioSource.stop();
+      audioSource = null;
     }
     if (audioContext) {
       audioContext.close();
       audioContext = null;
     }
   }
+  function close() {
+    if (ws) {
+      ws.close();
+      ws = null;
+    }
+    stopAudio();
+  }
   return {
     connect,
     speak,
     close,
-    play: playAudio
+    play: playAudio,
+    stop: stopAudio
   };
 }
+var tts_client_default = (authConfig) => {
+  return (config) => {
+    return createTTSClient({ ...authConfig, ...config });
+  };
+};
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {
-  createTTSClient,
-  createTtsClient
+  createTTSClient
 });
 //# sourceMappingURL=index.cjs.map

package/dist/index.cjs.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/index.ts","../src/tts-client.ts"],"sourcesContent":["export * from './tts-client';\nexport { createTTSClient as createTtsClient } from './tts-client';\n","/**\n * TTS Realtime WebSocket Client\n */\n\nconst TTS_PATH = \"/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime\";\n\nexport interface TTSClientConfig {\n  path?: string;\n  /** Voice name, default 'Cherry' */\n  voice?: string;\n  /** Auto play audio, default true */\n  autoPlay?: boolean;\n  /** Audio format, default 'pcm' */\n  audioFormat?: \"pcm\" | \"mp3\" | \"wav\" | \"opus\";\n  /** Sample rate, default 24000 */\n  sampleRate?: number;\n  /** Called when connection is ready */\n  onReady?: () => void;\n  /** Called when audio playback starts */\n  onAudioStart?: () => void;\n  /** Called when audio playback ends */\n  onAudioEnd?: () => void;\n  /** Called on each audio chunk received */\n  onAudioChunk?: (chunk: string[]) => void;\n  /** Called on error */\n  onError?: (error: Error) => void;\n}\n\nexport interface TTSClient {\n  /** Connect to TTS service */\n  connect(): Promise<void>;\n  /** Synthesize speech from text */\n  speak(text: string): Promise<void>;\n  /** Play audio from chunks */\n  play(): void;\n  /** Close connection */\n  close(): void;\n}\n\nexport function createTTSClient(config: TTSClientConfig): TTSClient {\n  const {\n    path = TTS_PATH,\n    voice = \"Cherry\",\n    autoPlay = true,\n    audioFormat = \"pcm\",\n    sampleRate = 24000,\n    onReady,\n    onAudioStart,\n    onAudioEnd,\n    onAudioChunk,\n    onError,\n  } = config;\n\n  let ws: WebSocket | null = null;\n  let audioChunks: string[] = [];\n  let audioContext: AudioContext | null = null;\n\n  async function connect(): Promise<void> {\n    return new Promise((resolve, reject) => {\n      ws = new WebSocket(path);\n\n      ws.onopen = () => {};\n\n      ws.onmessage = (event) => {\n        const data = JSON.parse(event.data);\n\n        if (data.type === \"session.created\") {\n          ws!.send(\n            JSON.stringify({\n              type: \"session.update\",\n              session: {\n                mode: \"server_commit\",\n                voice,\n                response_format: audioFormat,\n                sample_rate: sampleRate,\n              },\n            })\n          );\n        }\n\n        if (data.type === \"session.updated\") {\n          onReady?.();\n          resolve();\n        }\n\n        if (data.type === \"response.audio.delta\") {\n          audioChunks.push(data.delta);\n          onAudioChunk?.(audioChunks);\n        }\n\n        if (data.type === \"response.audio.done\") {\n          onAudioChunk?.(audioChunks);\n          if (autoPlay && typeof window !== \"undefined\") {\n            playAudio();\n          }\n        }\n\n        if (data.type === \"response.done\") {\n          ws!.send(JSON.stringify({ type: \"session.finish\" }));\n        }\n\n        if (data.type === \"error\") {\n          const err = new Error(data.error?.message || \"Unknown error\");\n          onError?.(err);\n          reject(err);\n        }\n      };\n\n      ws.onerror = () => {\n        const err = new Error(\"WebSocket connection error\");\n        onError?.(err);\n        reject(err);\n      };\n\n      ws.onclose = () => {\n        ws = null;\n      };\n    });\n  }\n\n  async function speak(text: string): Promise<void> {\n    if (!ws || ws.readyState !== WebSocket.OPEN) {\n      throw new Error(\"WebSocket not connected\");\n    }\n\n    audioChunks = [];\n\n    ws.send(\n      JSON.stringify({\n        type: \"input_text_buffer.append\",\n        text,\n      })\n    );\n\n    setTimeout(() => {\n      ws!.send(\n        JSON.stringify({\n          type: \"input_text_buffer.commit\",\n        })\n      );\n    }, 100);\n  }\n\n  function playAudio() {\n    let chunks: string[] = audioChunks;\n    if (typeof window === \"undefined\") return;\n\n    try {\n      if (!audioContext) {\n        audioContext = new AudioContext();\n      }\n\n      onAudioStart?.();\n\n      let totalBytes = 0;\n      const allBytes: Uint8Array[] = [];\n\n      for (const chunk of chunks) {\n        const binaryString = atob(chunk);\n        const bytes = new Uint8Array(binaryString.length);\n        for (let i = 0; i < binaryString.length; i++) {\n          bytes[i] = binaryString.charCodeAt(i);\n        }\n        allBytes.push(bytes);\n        totalBytes += bytes.length;\n      }\n\n      const combined = new Uint8Array(totalBytes);\n      let offset = 0;\n      for (const bytes of allBytes) {\n        combined.set(bytes, offset);\n        offset += bytes.length;\n      }\n\n      const numSamples = combined.length / 2;\n      const audioBuffer = audioContext.createBuffer(1, numSamples, sampleRate);\n      const channelData = audioBuffer.getChannelData(0);\n\n      const dataView = new DataView(combined.buffer);\n      for (let i = 0; i < numSamples; i++) {\n        const int16 = dataView.getInt16(i * 2, true);\n        channelData[i] = int16 / 32768.0;\n      }\n\n      const source = audioContext.createBufferSource();\n      source.buffer = audioBuffer;\n      source.connect(audioContext.destination);\n      source.onended = () => onAudioEnd?.();\n      source.start(0);\n    } catch (err) {\n      onError?.(err as Error);\n    }\n  }\n\n  function close() {\n    if (ws) {\n      ws.close();\n      ws = null;\n    }\n    if (audioContext) {\n      audioContext.close();\n      audioContext = null;\n    }\n  }\n\n  return {\n    connect,\n    speak,\n    close,\n    play: playAudio,\n  };\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;;;ACIA,IAAM,WAAW;AAmCV,SAAS,gBAAgB,QAAoC;AAClE,QAAM;AAAA,IACJ,OAAO;AAAA,IACP,QAAQ;AAAA,IACR,WAAW;AAAA,IACX,cAAc;AAAA,IACd,aAAa;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,IAAI;AAEJ,MAAI,KAAuB;AAC3B,MAAI,cAAwB,CAAC;AAC7B,MAAI,eAAoC;AAExC,iBAAe,UAAyB;AACtC,WAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AACtC,WAAK,IAAI,UAAU,IAAI;AAEvB,SAAG,SAAS,MAAM;AAAA,MAAC;AAEnB,SAAG,YAAY,CAAC,UAAU;AACxB,cAAM,OAAO,KAAK,MAAM,MAAM,IAAI;AAElC,YAAI,KAAK,SAAS,mBAAmB;AACnC,aAAI;AAAA,YACF,KAAK,UAAU;AAAA,cACb,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,MAAM;AAAA,gBACN;AAAA,gBACA,iBAAiB;AAAA,gBACjB,aAAa;AAAA,cACf;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,mBAAmB;AACnC,oBAAU;AACV,kBAAQ;AAAA,QACV;AAEA,YAAI,KAAK,SAAS,wBAAwB;AACxC,sBAAY,KAAK,KAAK,KAAK;AAC3B,yBAAe,WAAW;AAAA,QAC5B;AAEA,YAAI,KAAK,SAAS,uBAAuB;AACvC,yBAAe,WAAW;AAC1B,cAAI,YAAY,OAAO,WAAW,aAAa;AAC7C,sBAAU;AAAA,UACZ;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,iBAAiB;AACjC,aAAI,KAAK,KAAK,UAAU,EAAE,MAAM,iBAAiB,CAAC,CAAC;AAAA,QACrD;AAEA,YAAI,KAAK,SAAS,SAAS;AACzB,gBAAM,MAAM,IAAI,MAAM,KAAK,OAAO,WAAW,eAAe;AAC5D,oBAAU,GAAG;AACb,iBAAO,GAAG;AAAA,QACZ;AAAA,MACF;AAEA,SAAG,UAAU,MAAM;AACjB,cAAM,MAAM,IAAI,MAAM,4BAA4B;AAClD,kBAAU,GAAG;AACb,eAAO,GAAG;AAAA,MACZ;AAEA,SAAG,UAAU,MAAM;AACjB,aAAK;AAAA,MACP;AAAA,IACF,CAAC;AAAA,EACH;AAEA,iBAAe,MAAM,MAA6B;AAChD,QAAI,CAAC,MAAM,GAAG,eAAe,UAAU,MAAM;AAC3C,YAAM,IAAI,MAAM,yBAAyB;AAAA,IAC3C;AAEA,kBAAc,CAAC;AAEf,OAAG;AAAA,MACD,KAAK,UAAU;AAAA,QACb,MAAM;AAAA,QACN;AAAA,MACF,CAAC;AAAA,IACH;AAEA,eAAW,MAAM;AACf,SAAI;AAAA,QACF,KAAK,UAAU;AAAA,UACb,MAAM;AAAA,QACR,CAAC;AAAA,MACH;AAAA,IACF,GAAG,GAAG;AAAA,EACR;AAEA,WAAS,YAAY;AACnB,QAAI,SAAmB;AACvB,QAAI,OAAO,WAAW,YAAa;AAEnC,QAAI;AACF,UAAI,CAAC,cAAc;AACjB,uBAAe,IAAI,aAAa;AAAA,MAClC;AAEA,qBAAe;AAEf,UAAI,aAAa;AACjB,YAAM,WAAyB,CAAC;AAEhC,iBAAW,SAAS,QAAQ;AAC1B,cAAM,eAAe,KAAK,KAAK;AAC/B,cAAM,QAAQ,IAAI,WAAW,aAAa,MAAM;AAChD,iBAAS,IAAI,GAAG,IAAI,aAAa,QAAQ,KAAK;AAC5C,gBAAM,CAAC,IAAI,aAAa,WAAW,CAAC;AAAA,QACtC;AACA,iBAAS,KAAK,KAAK;AACnB,sBAAc,MAAM;AAAA,MACtB;AAEA,YAAM,WAAW,IAAI,WAAW,UAAU;AAC1C,UAAI,SAAS;AACb,iBAAW,SAAS,UAAU;AAC5B,iBAAS,IAAI,OAAO,MAAM;AAC1B,kBAAU,MAAM;AAAA,MAClB;AAEA,YAAM,aAAa,SAAS,SAAS;AACrC,YAAM,cAAc,aAAa,aAAa,GAAG,YAAY,UAAU;AACvE,YAAM,cAAc,YAAY,eAAe,CAAC;AAEhD,YAAM,WAAW,IAAI,SAAS,SAAS,MAAM;AAC7C,eAAS,IAAI,GAAG,IAAI,YAAY,KAAK;AACnC,cAAM,QAAQ,SAAS,SAAS,IAAI,GAAG,IAAI;AAC3C,oBAAY,CAAC,IAAI,QAAQ;AAAA,MAC3B;AAEA,YAAM,SAAS,aAAa,mBAAmB;AAC/C,aAAO,SAAS;AAChB,aAAO,QAAQ,aAAa,WAAW;AACvC,aAAO,UAAU,MAAM,aAAa;AACpC,aAAO,MAAM,CAAC;AAAA,IAChB,SAAS,KAAK;AACZ,gBAAU,GAAY;AAAA,IACxB;AAAA,EACF;AAEA,WAAS,QAAQ;AACf,QAAI,IAAI;AACN,SAAG,MAAM;AACT,WAAK;AAAA,IACP;AACA,QAAI,cAAc;AAChB,mBAAa,MAAM;AACnB,qBAAe;AAAA,IACjB;AAAA,EACF;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA,MAAM;AAAA,EACR;AACF;","names":[]}
+{"version":3,"sources":["../src/index.ts","../src/tts-client.ts"],"sourcesContent":["import type { TTSClient, TTSClientConfig } from \"./tts-client\";\nimport createTTSClient from \"./tts-client\";\nexport { createTTSClient, type TTSClient, type TTSClientConfig };\n","/**\n * TTS Realtime WebSocket Client\n *\n * WebSocket-based real-time text-to-speech synthesis with multiple voice options.\n * Built-in playback only supports PCM format.\n *\n * @example\n * ```typescript\n * const client = createTTSClient({\n *   voice: \"Cherry\",\n *   autoPlay: true,\n *   onReady() {\n *     console.log(\"TTS ready\");\n *   },\n *   onAudioStart() {\n *     console.log(\"Playing audio\");\n *   },\n *   onAudioEnd() {\n *     console.log(\"Playback ended\");\n *   },\n * });\n *\n * await client.connect();\n * await client.speak(\"Hello, this is a test.\");\n * // client.close();\n * ```\n */\n\nconst TTS_PATH = \"/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime\";\n\nexport interface TTSClientConfig {\n  /** Get access token for WebSocket authentication */\n  getAccessToken?: () => string | null;\n  /** Voice name, default 'Cherry' */\n  voice?: string;\n  /** Auto play audio, default true */\n  autoPlay?: boolean;\n  /** Audio format, default 'pcm' */\n  audioFormat?: \"pcm\" | \"mp3\" | \"wav\" | \"opus\";\n  /** Sample rate, default 24000 */\n  sampleRate?: number;\n  /** Called when connection is ready */\n  onReady?: () => void;\n  /** Called when audio playback starts */\n  onAudioStart?: () => void;\n  /** Called when audio playback ends */\n  onAudioEnd?: () => void;\n  /** Called on each audio chunk received */\n  onAudioChunk?: (chunk: string[]) => void;\n  /** Called on error */\n  onError?: (error: Error) => void;\n}\n\nexport interface TTSClient {\n  /** Connect to TTS service */\n  connect(): Promise<void>;\n  /** Synthesize speech from text */\n  speak(text: string): Promise<void>;\n  /** Play audio from chunks */\n  play(): void;\n  /** Stop audio playback */\n  stop(): void;\n  /** Close connection */\n  close(): void;\n}\n\nfunction createTTSClient(config: TTSClientConfig): TTSClient {\n  const {\n    getAccessToken,\n    voice = \"Cherry\",\n    autoPlay = true,\n    audioFormat = \"pcm\",\n    sampleRate = 24000,\n    onReady,\n    onAudioStart,\n    onAudioEnd,\n    onAudioChunk,\n    onError,\n  } = config;\n\n  let ws: WebSocket | null = null;\n  let audioChunks: string[] = [];\n  let audioContext: AudioContext | null = null;\n  let audioSource: AudioBufferSourceNode | null = null;\n\n  async function connect(): Promise<void> {\n    return new Promise((resolve, reject) => {\n      // Build WebSocket URL with optional token parameter\n      let wsUrl = TTS_PATH;\n      if (getAccessToken) {\n        const token = getAccessToken();\n        if (token) {\n          const separator = wsUrl.includes(\"?\") ? \"&\" : \"?\";\n          wsUrl = `${wsUrl}${separator}token=${encodeURIComponent(token)}`;\n        }\n      }\n\n      ws = new WebSocket(wsUrl);\n\n      ws.onopen = () => {};\n\n      ws.onmessage = (event) => {\n        const data = JSON.parse(event.data);\n\n        if (data.type === \"session.created\") {\n          ws!.send(\n            JSON.stringify({\n              type: \"session.update\",\n              session: {\n                mode: \"server_commit\",\n                voice,\n                response_format: audioFormat,\n                sample_rate: sampleRate,\n              },\n            })\n          );\n        }\n\n        if (data.type === \"session.updated\") {\n          onReady?.();\n          resolve();\n        }\n\n        if (data.type === \"response.audio.delta\") {\n          audioChunks.push(data.delta);\n          onAudioChunk?.(audioChunks);\n        }\n\n        if (data.type === \"response.audio.done\") {\n          onAudioChunk?.(audioChunks);\n          if (autoPlay && typeof window !== \"undefined\") {\n            playAudio();\n          }\n        }\n\n        if (data.type === \"response.done\") {\n          ws!.send(JSON.stringify({ type: \"session.finish\" }));\n        }\n\n        if (data.type === \"error\") {\n          const err = new Error(data.error?.message || \"Unknown error\");\n          onError?.(err);\n          reject(err);\n        }\n      };\n\n      ws.onerror = () => {\n        const err = new Error(\"WebSocket connection error\");\n        onError?.(err);\n        reject(err);\n      };\n\n      ws.onclose = () => {\n        ws = null;\n      };\n    });\n  }\n\n  async function speak(text: string): Promise<void> {\n    if (!ws || ws.readyState !== WebSocket.OPEN) {\n      throw new Error(\"WebSocket not connected\");\n    }\n\n    audioChunks = [];\n\n    ws.send(\n      JSON.stringify({\n        type: \"input_text_buffer.append\",\n        text,\n      })\n    );\n\n    setTimeout(() => {\n      ws!.send(\n        JSON.stringify({\n          type: \"input_text_buffer.commit\",\n        })\n      );\n    }, 100);\n  }\n\n  function playAudio() {\n    let chunks: string[] = audioChunks;\n    if (typeof window === \"undefined\") return;\n\n    try {\n      if (!audioContext) {\n        audioContext = new AudioContext();\n      }\n\n      onAudioStart?.();\n\n      let totalBytes = 0;\n      const allBytes: Uint8Array[] = [];\n\n      for (const chunk of chunks) {\n        const binaryString = atob(chunk);\n        const bytes = new Uint8Array(binaryString.length);\n        for (let i = 0; i < binaryString.length; i++) {\n          bytes[i] = binaryString.charCodeAt(i);\n        }\n        allBytes.push(bytes);\n        totalBytes += bytes.length;\n      }\n\n      const combined = new Uint8Array(totalBytes);\n      let offset = 0;\n      for (const bytes of allBytes) {\n        combined.set(bytes, offset);\n        offset += bytes.length;\n      }\n\n      const numSamples = combined.length / 2;\n      const audioBuffer = audioContext.createBuffer(1, numSamples, sampleRate);\n      const channelData = audioBuffer.getChannelData(0);\n\n      const dataView = new DataView(combined.buffer);\n      for (let i = 0; i < numSamples; i++) {\n        const int16 = dataView.getInt16(i * 2, true);\n        channelData[i] = int16 / 32768.0;\n      }\n\n      const source = audioContext.createBufferSource();\n      source.buffer = audioBuffer;\n      source.connect(audioContext.destination);\n      source.onended = () => onAudioEnd?.();\n      source.start(0);\n      audioSource = source;\n    } catch (err) {\n      onError?.(err as Error);\n    }\n  }\n\n  function stopAudio() {\n    if (audioSource) {\n      audioSource.stop();\n      audioSource = null;\n    }\n    if (audioContext) {\n      audioContext.close();\n      audioContext = null;\n    }\n  }\n\n  function close() {\n    if (ws) {\n      ws.close();\n      ws = null;\n    }\n    stopAudio();\n  }\n\n  return {\n    connect,\n    speak,\n    close,\n    play: playAudio,\n    stop: stopAudio,\n  };\n}\n\nexport default (\n  authConfig: Pick<TTSClientConfig, \"getAccessToken\">\n): ((config: TTSClientConfig) => TTSClient) => {\n  return (config: TTSClientConfig) => {\n    return createTTSClient({ ...authConfig, ...config });\n  };\n};\n"],"mappings":";;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;;;AC4BA,IAAM,WAAW;AAsCjB,SAAS,gBAAgB,QAAoC;AAC3D,QAAM;AAAA,IACJ;AAAA,IACA,QAAQ;AAAA,IACR,WAAW;AAAA,IACX,cAAc;AAAA,IACd,aAAa;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,IAAI;AAEJ,MAAI,KAAuB;AAC3B,MAAI,cAAwB,CAAC;AAC7B,MAAI,eAAoC;AACxC,MAAI,cAA4C;AAEhD,iBAAe,UAAyB;AACtC,WAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AAEtC,UAAI,QAAQ;AACZ,UAAI,gBAAgB;AAClB,cAAM,QAAQ,eAAe;AAC7B,YAAI,OAAO;AACT,gBAAM,YAAY,MAAM,SAAS,GAAG,IAAI,MAAM;AAC9C,kBAAQ,GAAG,KAAK,GAAG,SAAS,SAAS,mBAAmB,KAAK,CAAC;AAAA,QAChE;AAAA,MACF;AAEA,WAAK,IAAI,UAAU,KAAK;AAExB,SAAG,SAAS,MAAM;AAAA,MAAC;AAEnB,SAAG,YAAY,CAAC,UAAU;AACxB,cAAM,OAAO,KAAK,MAAM,MAAM,IAAI;AAElC,YAAI,KAAK,SAAS,mBAAmB;AACnC,aAAI;AAAA,YACF,KAAK,UAAU;AAAA,cACb,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,MAAM;AAAA,gBACN;AAAA,gBACA,iBAAiB;AAAA,gBACjB,aAAa;AAAA,cACf;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,mBAAmB;AACnC,oBAAU;AACV,kBAAQ;AAAA,QACV;AAEA,YAAI,KAAK,SAAS,wBAAwB;AACxC,sBAAY,KAAK,KAAK,KAAK;AAC3B,yBAAe,WAAW;AAAA,QAC5B;AAEA,YAAI,KAAK,SAAS,uBAAuB;AACvC,yBAAe,WAAW;AAC1B,cAAI,YAAY,OAAO,WAAW,aAAa;AAC7C,sBAAU;AAAA,UACZ;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,iBAAiB;AACjC,aAAI,KAAK,KAAK,UAAU,EAAE,MAAM,iBAAiB,CAAC,CAAC;AAAA,QACrD;AAEA,YAAI,KAAK,SAAS,SAAS;AACzB,gBAAM,MAAM,IAAI,MAAM,KAAK,OAAO,WAAW,eAAe;AAC5D,oBAAU,GAAG;AACb,iBAAO,GAAG;AAAA,QACZ;AAAA,MACF;AAEA,SAAG,UAAU,MAAM;AACjB,cAAM,MAAM,IAAI,MAAM,4BAA4B;AAClD,kBAAU,GAAG;AACb,eAAO,GAAG;AAAA,MACZ;AAEA,SAAG,UAAU,MAAM;AACjB,aAAK;AAAA,MACP;AAAA,IACF,CAAC;AAAA,EACH;AAEA,iBAAe,MAAM,MAA6B;AAChD,QAAI,CAAC,MAAM,GAAG,eAAe,UAAU,MAAM;AAC3C,YAAM,IAAI,MAAM,yBAAyB;AAAA,IAC3C;AAEA,kBAAc,CAAC;AAEf,OAAG;AAAA,MACD,KAAK,UAAU;AAAA,QACb,MAAM;AAAA,QACN;AAAA,MACF,CAAC;AAAA,IACH;AAEA,eAAW,MAAM;AACf,SAAI;AAAA,QACF,KAAK,UAAU;AAAA,UACb,MAAM;AAAA,QACR,CAAC;AAAA,MACH;AAAA,IACF,GAAG,GAAG;AAAA,EACR;AAEA,WAAS,YAAY;AACnB,QAAI,SAAmB;AACvB,QAAI,OAAO,WAAW,YAAa;AAEnC,QAAI;AACF,UAAI,CAAC,cAAc;AACjB,uBAAe,IAAI,aAAa;AAAA,MAClC;AAEA,qBAAe;AAEf,UAAI,aAAa;AACjB,YAAM,WAAyB,CAAC;AAEhC,iBAAW,SAAS,QAAQ;AAC1B,cAAM,eAAe,KAAK,KAAK;AAC/B,cAAM,QAAQ,IAAI,WAAW,aAAa,MAAM;AAChD,iBAAS,IAAI,GAAG,IAAI,aAAa,QAAQ,KAAK;AAC5C,gBAAM,CAAC,IAAI,aAAa,WAAW,CAAC;AAAA,QACtC;AACA,iBAAS,KAAK,KAAK;AACnB,sBAAc,MAAM;AAAA,MACtB;AAEA,YAAM,WAAW,IAAI,WAAW,UAAU;AAC1C,UAAI,SAAS;AACb,iBAAW,SAAS,UAAU;AAC5B,iBAAS,IAAI,OAAO,MAAM;AAC1B,kBAAU,MAAM;AAAA,MAClB;AAEA,YAAM,aAAa,SAAS,SAAS;AACrC,YAAM,cAAc,aAAa,aAAa,GAAG,YAAY,UAAU;AACvE,YAAM,cAAc,YAAY,eAAe,CAAC;AAEhD,YAAM,WAAW,IAAI,SAAS,SAAS,MAAM;AAC7C,eAAS,IAAI,GAAG,IAAI,YAAY,KAAK;AACnC,cAAM,QAAQ,SAAS,SAAS,IAAI,GAAG,IAAI;AAC3C,oBAAY,CAAC,IAAI,QAAQ;AAAA,MAC3B;AAEA,YAAM,SAAS,aAAa,mBAAmB;AAC/C,aAAO,SAAS;AAChB,aAAO,QAAQ,aAAa,WAAW;AACvC,aAAO,UAAU,MAAM,aAAa;AACpC,aAAO,MAAM,CAAC;AACd,oBAAc;AAAA,IAChB,SAAS,KAAK;AACZ,gBAAU,GAAY;AAAA,IACxB;AAAA,EACF;AAEA,WAAS,YAAY;AACnB,QAAI,aAAa;AACf,kBAAY,KAAK;AACjB,oBAAc;AAAA,IAChB;AACA,QAAI,cAAc;AAChB,mBAAa,MAAM;AACnB,qBAAe;AAAA,IACjB;AAAA,EACF;AAEA,WAAS,QAAQ;AACf,QAAI,IAAI;AACN,SAAG,MAAM;AACT,WAAK;AAAA,IACP;AACA,cAAU;AAAA,EACZ;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA,MAAM;AAAA,IACN,MAAM;AAAA,EACR;AACF;AAEA,IAAO,qBAAQ,CACb,eAC6C;AAC7C,SAAO,CAAC,WAA4B;AAClC,WAAO,gBAAgB,EAAE,GAAG,YAAY,GAAG,OAAO,CAAC;AAAA,EACrD;AACF;","names":[]}

package/dist/index.d.cts

@@ -1,8 +1,33 @@
 /**
  * TTS Realtime WebSocket Client
+ *
+ * WebSocket-based real-time text-to-speech synthesis with multiple voice options.
+ * Built-in playback only supports PCM format.
+ *
+ * @example
+ * ```typescript
+ * const client = createTTSClient({
+ *   voice: "Cherry",
+ *   autoPlay: true,
+ *   onReady() {
+ *     console.log("TTS ready");
+ *   },
+ *   onAudioStart() {
+ *     console.log("Playing audio");
+ *   },
+ *   onAudioEnd() {
+ *     console.log("Playback ended");
+ *   },
+ * });
+ *
+ * await client.connect();
+ * await client.speak("Hello, this is a test.");
+ * // client.close();
+ * ```
  */
 interface TTSClientConfig {
-    path?: string;
+    /** Get access token for WebSocket authentication */
+    getAccessToken?: () => string | null;
     /** Voice name, default 'Cherry' */
     voice?: string;
     /** Auto play audio, default true */
@@ -29,9 +54,11 @@ interface TTSClient {
     speak(text: string): Promise<void>;
     /** Play audio from chunks */
     play(): void;
+    /** Stop audio playback */
+    stop(): void;
     /** Close connection */
     close(): void;
 }
-declare function createTTSClient(config: TTSClientConfig): TTSClient;
+declare const _default: (authConfig: Pick<TTSClientConfig, "getAccessToken">) => ((config: TTSClientConfig) => TTSClient);
 
-export { type TTSClient, type TTSClientConfig, createTTSClient, createTTSClient as createTtsClient };
+export { type TTSClient, type TTSClientConfig, _default as createTTSClient };

package/dist/index.d.ts

@@ -1,8 +1,33 @@
 /**
  * TTS Realtime WebSocket Client
+ *
+ * WebSocket-based real-time text-to-speech synthesis with multiple voice options.
+ * Built-in playback only supports PCM format.
+ *
+ * @example
+ * ```typescript
+ * const client = createTTSClient({
+ *   voice: "Cherry",
+ *   autoPlay: true,
+ *   onReady() {
+ *     console.log("TTS ready");
+ *   },
+ *   onAudioStart() {
+ *     console.log("Playing audio");
+ *   },
+ *   onAudioEnd() {
+ *     console.log("Playback ended");
+ *   },
+ * });
+ *
+ * await client.connect();
+ * await client.speak("Hello, this is a test.");
+ * // client.close();
+ * ```
  */
 interface TTSClientConfig {
-    path?: string;
+    /** Get access token for WebSocket authentication */
+    getAccessToken?: () => string | null;
     /** Voice name, default 'Cherry' */
     voice?: string;
     /** Auto play audio, default true */
@@ -29,9 +54,11 @@ interface TTSClient {
     speak(text: string): Promise<void>;
     /** Play audio from chunks */
     play(): void;
+    /** Stop audio playback */
+    stop(): void;
     /** Close connection */
     close(): void;
 }
-declare function createTTSClient(config: TTSClientConfig): TTSClient;
+declare const _default: (authConfig: Pick<TTSClientConfig, "getAccessToken">) => ((config: TTSClientConfig) => TTSClient);
 
-export { type TTSClient, type TTSClientConfig, createTTSClient, createTTSClient as createTtsClient };
+export { type TTSClient, type TTSClientConfig, _default as createTTSClient };

package/dist/index.js

@@ -2,7 +2,7 @@
 var TTS_PATH = "/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime";
 function createTTSClient(config) {
   const {
-    path = TTS_PATH,
+    getAccessToken,
     voice = "Cherry",
     autoPlay = true,
     audioFormat = "pcm",
@@ -16,9 +16,18 @@ function createTTSClient(config) {
   let ws = null;
   let audioChunks = [];
   let audioContext = null;
+  let audioSource = null;
   async function connect() {
     return new Promise((resolve, reject) => {
-      ws = new WebSocket(path);
+      let wsUrl = TTS_PATH;
+      if (getAccessToken) {
+        const token = getAccessToken();
+        if (token) {
+          const separator = wsUrl.includes("?") ? "&" : "?";
+          wsUrl = `${wsUrl}${separator}token=${encodeURIComponent(token)}`;
+        }
+      }
+      ws = new WebSocket(wsUrl);
       ws.onopen = () => {
       };
       ws.onmessage = (event) => {
@@ -126,29 +135,42 @@ function createTTSClient(config) {
       source.connect(audioContext.destination);
       source.onended = () => onAudioEnd?.();
       source.start(0);
+      audioSource = source;
     } catch (err) {
       onError?.(err);
     }
   }
-  function close() {
-    if (ws) {
-      ws.close();
-      ws = null;
+  function stopAudio() {
+    if (audioSource) {
+      audioSource.stop();
+      audioSource = null;
     }
     if (audioContext) {
       audioContext.close();
       audioContext = null;
     }
   }
+  function close() {
+    if (ws) {
+      ws.close();
+      ws = null;
+    }
+    stopAudio();
+  }
   return {
     connect,
     speak,
     close,
-    play: playAudio
+    play: playAudio,
+    stop: stopAudio
   };
 }
+var tts_client_default = (authConfig) => {
+  return (config) => {
+    return createTTSClient({ ...authConfig, ...config });
+  };
+};
 export {
-  createTTSClient,
-  createTTSClient as createTtsClient
+  tts_client_default as createTTSClient
 };
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/tts-client.ts"],"sourcesContent":["/**\n * TTS Realtime WebSocket Client\n */\n\nconst TTS_PATH = \"/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime\";\n\nexport interface TTSClientConfig {\n  path?: string;\n  /** Voice name, default 'Cherry' */\n  voice?: string;\n  /** Auto play audio, default true */\n  autoPlay?: boolean;\n  /** Audio format, default 'pcm' */\n  audioFormat?: \"pcm\" | \"mp3\" | \"wav\" | \"opus\";\n  /** Sample rate, default 24000 */\n  sampleRate?: number;\n  /** Called when connection is ready */\n  onReady?: () => void;\n  /** Called when audio playback starts */\n  onAudioStart?: () => void;\n  /** Called when audio playback ends */\n  onAudioEnd?: () => void;\n  /** Called on each audio chunk received */\n  onAudioChunk?: (chunk: string[]) => void;\n  /** Called on error */\n  onError?: (error: Error) => void;\n}\n\nexport interface TTSClient {\n  /** Connect to TTS service */\n  connect(): Promise<void>;\n  /** Synthesize speech from text */\n  speak(text: string): Promise<void>;\n  /** Play audio from chunks */\n  play(): void;\n  /** Close connection */\n  close(): void;\n}\n\nexport function createTTSClient(config: TTSClientConfig): TTSClient {\n  const {\n    path = TTS_PATH,\n    voice = \"Cherry\",\n    autoPlay = true,\n    audioFormat = \"pcm\",\n    sampleRate = 24000,\n    onReady,\n    onAudioStart,\n    onAudioEnd,\n    onAudioChunk,\n    onError,\n  } = config;\n\n  let ws: WebSocket | null = null;\n  let audioChunks: string[] = [];\n  let audioContext: AudioContext | null = null;\n\n  async function connect(): Promise<void> {\n    return new Promise((resolve, reject) => {\n      ws = new WebSocket(path);\n\n      ws.onopen = () => {};\n\n      ws.onmessage = (event) => {\n        const data = JSON.parse(event.data);\n\n        if (data.type === \"session.created\") {\n          ws!.send(\n            JSON.stringify({\n              type: \"session.update\",\n              session: {\n                mode: \"server_commit\",\n                voice,\n                response_format: audioFormat,\n                sample_rate: sampleRate,\n              },\n            })\n          );\n        }\n\n        if (data.type === \"session.updated\") {\n          onReady?.();\n          resolve();\n        }\n\n        if (data.type === \"response.audio.delta\") {\n          audioChunks.push(data.delta);\n          onAudioChunk?.(audioChunks);\n        }\n\n        if (data.type === \"response.audio.done\") {\n          onAudioChunk?.(audioChunks);\n          if (autoPlay && typeof window !== \"undefined\") {\n            playAudio();\n          }\n        }\n\n        if (data.type === \"response.done\") {\n          ws!.send(JSON.stringify({ type: \"session.finish\" }));\n        }\n\n        if (data.type === \"error\") {\n          const err = new Error(data.error?.message || \"Unknown error\");\n          onError?.(err);\n          reject(err);\n        }\n      };\n\n      ws.onerror = () => {\n        const err = new Error(\"WebSocket connection error\");\n        onError?.(err);\n        reject(err);\n      };\n\n      ws.onclose = () => {\n        ws = null;\n      };\n    });\n  }\n\n  async function speak(text: string): Promise<void> {\n    if (!ws || ws.readyState !== WebSocket.OPEN) {\n      throw new Error(\"WebSocket not connected\");\n    }\n\n    audioChunks = [];\n\n    ws.send(\n      JSON.stringify({\n        type: \"input_text_buffer.append\",\n        text,\n      })\n    );\n\n    setTimeout(() => {\n      ws!.send(\n        JSON.stringify({\n          type: \"input_text_buffer.commit\",\n        })\n      );\n    }, 100);\n  }\n\n  function playAudio() {\n    let chunks: string[] = audioChunks;\n    if (typeof window === \"undefined\") return;\n\n    try {\n      if (!audioContext) {\n        audioContext = new AudioContext();\n      }\n\n      onAudioStart?.();\n\n      let totalBytes = 0;\n      const allBytes: Uint8Array[] = [];\n\n      for (const chunk of chunks) {\n        const binaryString = atob(chunk);\n        const bytes = new Uint8Array(binaryString.length);\n        for (let i = 0; i < binaryString.length; i++) {\n          bytes[i] = binaryString.charCodeAt(i);\n        }\n        allBytes.push(bytes);\n        totalBytes += bytes.length;\n      }\n\n      const combined = new Uint8Array(totalBytes);\n      let offset = 0;\n      for (const bytes of allBytes) {\n        combined.set(bytes, offset);\n        offset += bytes.length;\n      }\n\n      const numSamples = combined.length / 2;\n      const audioBuffer = audioContext.createBuffer(1, numSamples, sampleRate);\n      const channelData = audioBuffer.getChannelData(0);\n\n      const dataView = new DataView(combined.buffer);\n      for (let i = 0; i < numSamples; i++) {\n        const int16 = dataView.getInt16(i * 2, true);\n        channelData[i] = int16 / 32768.0;\n      }\n\n      const source = audioContext.createBufferSource();\n      source.buffer = audioBuffer;\n      source.connect(audioContext.destination);\n      source.onended = () => onAudioEnd?.();\n      source.start(0);\n    } catch (err) {\n      onError?.(err as Error);\n    }\n  }\n\n  function close() {\n    if (ws) {\n      ws.close();\n      ws = null;\n    }\n    if (audioContext) {\n      audioContext.close();\n      audioContext = null;\n    }\n  }\n\n  return {\n    connect,\n    speak,\n    close,\n    play: playAudio,\n  };\n}\n"],"mappings":";AAIA,IAAM,WAAW;AAmCV,SAAS,gBAAgB,QAAoC;AAClE,QAAM;AAAA,IACJ,OAAO;AAAA,IACP,QAAQ;AAAA,IACR,WAAW;AAAA,IACX,cAAc;AAAA,IACd,aAAa;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,IAAI;AAEJ,MAAI,KAAuB;AAC3B,MAAI,cAAwB,CAAC;AAC7B,MAAI,eAAoC;AAExC,iBAAe,UAAyB;AACtC,WAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AACtC,WAAK,IAAI,UAAU,IAAI;AAEvB,SAAG,SAAS,MAAM;AAAA,MAAC;AAEnB,SAAG,YAAY,CAAC,UAAU;AACxB,cAAM,OAAO,KAAK,MAAM,MAAM,IAAI;AAElC,YAAI,KAAK,SAAS,mBAAmB;AACnC,aAAI;AAAA,YACF,KAAK,UAAU;AAAA,cACb,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,MAAM;AAAA,gBACN;AAAA,gBACA,iBAAiB;AAAA,gBACjB,aAAa;AAAA,cACf;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,mBAAmB;AACnC,oBAAU;AACV,kBAAQ;AAAA,QACV;AAEA,YAAI,KAAK,SAAS,wBAAwB;AACxC,sBAAY,KAAK,KAAK,KAAK;AAC3B,yBAAe,WAAW;AAAA,QAC5B;AAEA,YAAI,KAAK,SAAS,uBAAuB;AACvC,yBAAe,WAAW;AAC1B,cAAI,YAAY,OAAO,WAAW,aAAa;AAC7C,sBAAU;AAAA,UACZ;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,iBAAiB;AACjC,aAAI,KAAK,KAAK,UAAU,EAAE,MAAM,iBAAiB,CAAC,CAAC;AAAA,QACrD;AAEA,YAAI,KAAK,SAAS,SAAS;AACzB,gBAAM,MAAM,IAAI,MAAM,KAAK,OAAO,WAAW,eAAe;AAC5D,oBAAU,GAAG;AACb,iBAAO,GAAG;AAAA,QACZ;AAAA,MACF;AAEA,SAAG,UAAU,MAAM;AACjB,cAAM,MAAM,IAAI,MAAM,4BAA4B;AAClD,kBAAU,GAAG;AACb,eAAO,GAAG;AAAA,MACZ;AAEA,SAAG,UAAU,MAAM;AACjB,aAAK;AAAA,MACP;AAAA,IACF,CAAC;AAAA,EACH;AAEA,iBAAe,MAAM,MAA6B;AAChD,QAAI,CAAC,MAAM,GAAG,eAAe,UAAU,MAAM;AAC3C,YAAM,IAAI,MAAM,yBAAyB;AAAA,IAC3C;AAEA,kBAAc,CAAC;AAEf,OAAG;AAAA,MACD,KAAK,UAAU;AAAA,QACb,MAAM;AAAA,QACN;AAAA,MACF,CAAC;AAAA,IACH;AAEA,eAAW,MAAM;AACf,SAAI;AAAA,QACF,KAAK,UAAU;AAAA,UACb,MAAM;AAAA,QACR,CAAC;AAAA,MACH;AAAA,IACF,GAAG,GAAG;AAAA,EACR;AAEA,WAAS,YAAY;AACnB,QAAI,SAAmB;AACvB,QAAI,OAAO,WAAW,YAAa;AAEnC,QAAI;AACF,UAAI,CAAC,cAAc;AACjB,uBAAe,IAAI,aAAa;AAAA,MAClC;AAEA,qBAAe;AAEf,UAAI,aAAa;AACjB,YAAM,WAAyB,CAAC;AAEhC,iBAAW,SAAS,QAAQ;AAC1B,cAAM,eAAe,KAAK,KAAK;AAC/B,cAAM,QAAQ,IAAI,WAAW,aAAa,MAAM;AAChD,iBAAS,IAAI,GAAG,IAAI,aAAa,QAAQ,KAAK;AAC5C,gBAAM,CAAC,IAAI,aAAa,WAAW,CAAC;AAAA,QACtC;AACA,iBAAS,KAAK,KAAK;AACnB,sBAAc,MAAM;AAAA,MACtB;AAEA,YAAM,WAAW,IAAI,WAAW,UAAU;AAC1C,UAAI,SAAS;AACb,iBAAW,SAAS,UAAU;AAC5B,iBAAS,IAAI,OAAO,MAAM;AAC1B,kBAAU,MAAM;AAAA,MAClB;AAEA,YAAM,aAAa,SAAS,SAAS;AACrC,YAAM,cAAc,aAAa,aAAa,GAAG,YAAY,UAAU;AACvE,YAAM,cAAc,YAAY,eAAe,CAAC;AAEhD,YAAM,WAAW,IAAI,SAAS,SAAS,MAAM;AAC7C,eAAS,IAAI,GAAG,IAAI,YAAY,KAAK;AACnC,cAAM,QAAQ,SAAS,SAAS,IAAI,GAAG,IAAI;AAC3C,oBAAY,CAAC,IAAI,QAAQ;AAAA,MAC3B;AAEA,YAAM,SAAS,aAAa,mBAAmB;AAC/C,aAAO,SAAS;AAChB,aAAO,QAAQ,aAAa,WAAW;AACvC,aAAO,UAAU,MAAM,aAAa;AACpC,aAAO,MAAM,CAAC;AAAA,IAChB,SAAS,KAAK;AACZ,gBAAU,GAAY;AAAA,IACxB;AAAA,EACF;AAEA,WAAS,QAAQ;AACf,QAAI,IAAI;AACN,SAAG,MAAM;AACT,WAAK;AAAA,IACP;AACA,QAAI,cAAc;AAChB,mBAAa,MAAM;AACnB,qBAAe;AAAA,IACjB;AAAA,EACF;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA,MAAM;AAAA,EACR;AACF;","names":[]}
+{"version":3,"sources":["../src/tts-client.ts"],"sourcesContent":["/**\n * TTS Realtime WebSocket Client\n *\n * WebSocket-based real-time text-to-speech synthesis with multiple voice options.\n * Built-in playback only supports PCM format.\n *\n * @example\n * ```typescript\n * const client = createTTSClient({\n *   voice: \"Cherry\",\n *   autoPlay: true,\n *   onReady() {\n *     console.log(\"TTS ready\");\n *   },\n *   onAudioStart() {\n *     console.log(\"Playing audio\");\n *   },\n *   onAudioEnd() {\n *     console.log(\"Playback ended\");\n *   },\n * });\n *\n * await client.connect();\n * await client.speak(\"Hello, this is a test.\");\n * // client.close();\n * ```\n */\n\nconst TTS_PATH = \"/api/proxy/builtin/platform/qwen-tts/api-ws/v1/realtime\";\n\nexport interface TTSClientConfig {\n  /** Get access token for WebSocket authentication */\n  getAccessToken?: () => string | null;\n  /** Voice name, default 'Cherry' */\n  voice?: string;\n  /** Auto play audio, default true */\n  autoPlay?: boolean;\n  /** Audio format, default 'pcm' */\n  audioFormat?: \"pcm\" | \"mp3\" | \"wav\" | \"opus\";\n  /** Sample rate, default 24000 */\n  sampleRate?: number;\n  /** Called when connection is ready */\n  onReady?: () => void;\n  /** Called when audio playback starts */\n  onAudioStart?: () => void;\n  /** Called when audio playback ends */\n  onAudioEnd?: () => void;\n  /** Called on each audio chunk received */\n  onAudioChunk?: (chunk: string[]) => void;\n  /** Called on error */\n  onError?: (error: Error) => void;\n}\n\nexport interface TTSClient {\n  /** Connect to TTS service */\n  connect(): Promise<void>;\n  /** Synthesize speech from text */\n  speak(text: string): Promise<void>;\n  /** Play audio from chunks */\n  play(): void;\n  /** Stop audio playback */\n  stop(): void;\n  /** Close connection */\n  close(): void;\n}\n\nfunction createTTSClient(config: TTSClientConfig): TTSClient {\n  const {\n    getAccessToken,\n    voice = \"Cherry\",\n    autoPlay = true,\n    audioFormat = \"pcm\",\n    sampleRate = 24000,\n    onReady,\n    onAudioStart,\n    onAudioEnd,\n    onAudioChunk,\n    onError,\n  } = config;\n\n  let ws: WebSocket | null = null;\n  let audioChunks: string[] = [];\n  let audioContext: AudioContext | null = null;\n  let audioSource: AudioBufferSourceNode | null = null;\n\n  async function connect(): Promise<void> {\n    return new Promise((resolve, reject) => {\n      // Build WebSocket URL with optional token parameter\n      let wsUrl = TTS_PATH;\n      if (getAccessToken) {\n        const token = getAccessToken();\n        if (token) {\n          const separator = wsUrl.includes(\"?\") ? \"&\" : \"?\";\n          wsUrl = `${wsUrl}${separator}token=${encodeURIComponent(token)}`;\n        }\n      }\n\n      ws = new WebSocket(wsUrl);\n\n      ws.onopen = () => {};\n\n      ws.onmessage = (event) => {\n        const data = JSON.parse(event.data);\n\n        if (data.type === \"session.created\") {\n          ws!.send(\n            JSON.stringify({\n              type: \"session.update\",\n              session: {\n                mode: \"server_commit\",\n                voice,\n                response_format: audioFormat,\n                sample_rate: sampleRate,\n              },\n            })\n          );\n        }\n\n        if (data.type === \"session.updated\") {\n          onReady?.();\n          resolve();\n        }\n\n        if (data.type === \"response.audio.delta\") {\n          audioChunks.push(data.delta);\n          onAudioChunk?.(audioChunks);\n        }\n\n        if (data.type === \"response.audio.done\") {\n          onAudioChunk?.(audioChunks);\n          if (autoPlay && typeof window !== \"undefined\") {\n            playAudio();\n          }\n        }\n\n        if (data.type === \"response.done\") {\n          ws!.send(JSON.stringify({ type: \"session.finish\" }));\n        }\n\n        if (data.type === \"error\") {\n          const err = new Error(data.error?.message || \"Unknown error\");\n          onError?.(err);\n          reject(err);\n        }\n      };\n\n      ws.onerror = () => {\n        const err = new Error(\"WebSocket connection error\");\n        onError?.(err);\n        reject(err);\n      };\n\n      ws.onclose = () => {\n        ws = null;\n      };\n    });\n  }\n\n  async function speak(text: string): Promise<void> {\n    if (!ws || ws.readyState !== WebSocket.OPEN) {\n      throw new Error(\"WebSocket not connected\");\n    }\n\n    audioChunks = [];\n\n    ws.send(\n      JSON.stringify({\n        type: \"input_text_buffer.append\",\n        text,\n      })\n    );\n\n    setTimeout(() => {\n      ws!.send(\n        JSON.stringify({\n          type: \"input_text_buffer.commit\",\n        })\n      );\n    }, 100);\n  }\n\n  function playAudio() {\n    let chunks: string[] = audioChunks;\n    if (typeof window === \"undefined\") return;\n\n    try {\n      if (!audioContext) {\n        audioContext = new AudioContext();\n      }\n\n      onAudioStart?.();\n\n      let totalBytes = 0;\n      const allBytes: Uint8Array[] = [];\n\n      for (const chunk of chunks) {\n        const binaryString = atob(chunk);\n        const bytes = new Uint8Array(binaryString.length);\n        for (let i = 0; i < binaryString.length; i++) {\n          bytes[i] = binaryString.charCodeAt(i);\n        }\n        allBytes.push(bytes);\n        totalBytes += bytes.length;\n      }\n\n      const combined = new Uint8Array(totalBytes);\n      let offset = 0;\n      for (const bytes of allBytes) {\n        combined.set(bytes, offset);\n        offset += bytes.length;\n      }\n\n      const numSamples = combined.length / 2;\n      const audioBuffer = audioContext.createBuffer(1, numSamples, sampleRate);\n      const channelData = audioBuffer.getChannelData(0);\n\n      const dataView = new DataView(combined.buffer);\n      for (let i = 0; i < numSamples; i++) {\n        const int16 = dataView.getInt16(i * 2, true);\n        channelData[i] = int16 / 32768.0;\n      }\n\n      const source = audioContext.createBufferSource();\n      source.buffer = audioBuffer;\n      source.connect(audioContext.destination);\n      source.onended = () => onAudioEnd?.();\n      source.start(0);\n      audioSource = source;\n    } catch (err) {\n      onError?.(err as Error);\n    }\n  }\n\n  function stopAudio() {\n    if (audioSource) {\n      audioSource.stop();\n      audioSource = null;\n    }\n    if (audioContext) {\n      audioContext.close();\n      audioContext = null;\n    }\n  }\n\n  function close() {\n    if (ws) {\n      ws.close();\n      ws = null;\n    }\n    stopAudio();\n  }\n\n  return {\n    connect,\n    speak,\n    close,\n    play: playAudio,\n    stop: stopAudio,\n  };\n}\n\nexport default (\n  authConfig: Pick<TTSClientConfig, \"getAccessToken\">\n): ((config: TTSClientConfig) => TTSClient) => {\n  return (config: TTSClientConfig) => {\n    return createTTSClient({ ...authConfig, ...config });\n  };\n};\n"],"mappings":";AA4BA,IAAM,WAAW;AAsCjB,SAAS,gBAAgB,QAAoC;AAC3D,QAAM;AAAA,IACJ;AAAA,IACA,QAAQ;AAAA,IACR,WAAW;AAAA,IACX,cAAc;AAAA,IACd,aAAa;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,IAAI;AAEJ,MAAI,KAAuB;AAC3B,MAAI,cAAwB,CAAC;AAC7B,MAAI,eAAoC;AACxC,MAAI,cAA4C;AAEhD,iBAAe,UAAyB;AACtC,WAAO,IAAI,QAAQ,CAAC,SAAS,WAAW;AAEtC,UAAI,QAAQ;AACZ,UAAI,gBAAgB;AAClB,cAAM,QAAQ,eAAe;AAC7B,YAAI,OAAO;AACT,gBAAM,YAAY,MAAM,SAAS,GAAG,IAAI,MAAM;AAC9C,kBAAQ,GAAG,KAAK,GAAG,SAAS,SAAS,mBAAmB,KAAK,CAAC;AAAA,QAChE;AAAA,MACF;AAEA,WAAK,IAAI,UAAU,KAAK;AAExB,SAAG,SAAS,MAAM;AAAA,MAAC;AAEnB,SAAG,YAAY,CAAC,UAAU;AACxB,cAAM,OAAO,KAAK,MAAM,MAAM,IAAI;AAElC,YAAI,KAAK,SAAS,mBAAmB;AACnC,aAAI;AAAA,YACF,KAAK,UAAU;AAAA,cACb,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,MAAM;AAAA,gBACN;AAAA,gBACA,iBAAiB;AAAA,gBACjB,aAAa;AAAA,cACf;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,mBAAmB;AACnC,oBAAU;AACV,kBAAQ;AAAA,QACV;AAEA,YAAI,KAAK,SAAS,wBAAwB;AACxC,sBAAY,KAAK,KAAK,KAAK;AAC3B,yBAAe,WAAW;AAAA,QAC5B;AAEA,YAAI,KAAK,SAAS,uBAAuB;AACvC,yBAAe,WAAW;AAC1B,cAAI,YAAY,OAAO,WAAW,aAAa;AAC7C,sBAAU;AAAA,UACZ;AAAA,QACF;AAEA,YAAI,KAAK,SAAS,iBAAiB;AACjC,aAAI,KAAK,KAAK,UAAU,EAAE,MAAM,iBAAiB,CAAC,CAAC;AAAA,QACrD;AAEA,YAAI,KAAK,SAAS,SAAS;AACzB,gBAAM,MAAM,IAAI,MAAM,KAAK,OAAO,WAAW,eAAe;AAC5D,oBAAU,GAAG;AACb,iBAAO,GAAG;AAAA,QACZ;AAAA,MACF;AAEA,SAAG,UAAU,MAAM;AACjB,cAAM,MAAM,IAAI,MAAM,4BAA4B;AAClD,kBAAU,GAAG;AACb,eAAO,GAAG;AAAA,MACZ;AAEA,SAAG,UAAU,MAAM;AACjB,aAAK;AAAA,MACP;AAAA,IACF,CAAC;AAAA,EACH;AAEA,iBAAe,MAAM,MAA6B;AAChD,QAAI,CAAC,MAAM,GAAG,eAAe,UAAU,MAAM;AAC3C,YAAM,IAAI,MAAM,yBAAyB;AAAA,IAC3C;AAEA,kBAAc,CAAC;AAEf,OAAG;AAAA,MACD,KAAK,UAAU;AAAA,QACb,MAAM;AAAA,QACN;AAAA,MACF,CAAC;AAAA,IACH;AAEA,eAAW,MAAM;AACf,SAAI;AAAA,QACF,KAAK,UAAU;AAAA,UACb,MAAM;AAAA,QACR,CAAC;AAAA,MACH;AAAA,IACF,GAAG,GAAG;AAAA,EACR;AAEA,WAAS,YAAY;AACnB,QAAI,SAAmB;AACvB,QAAI,OAAO,WAAW,YAAa;AAEnC,QAAI;AACF,UAAI,CAAC,cAAc;AACjB,uBAAe,IAAI,aAAa;AAAA,MAClC;AAEA,qBAAe;AAEf,UAAI,aAAa;AACjB,YAAM,WAAyB,CAAC;AAEhC,iBAAW,SAAS,QAAQ;AAC1B,cAAM,eAAe,KAAK,KAAK;AAC/B,cAAM,QAAQ,IAAI,WAAW,aAAa,MAAM;AAChD,iBAAS,IAAI,GAAG,IAAI,aAAa,QAAQ,KAAK;AAC5C,gBAAM,CAAC,IAAI,aAAa,WAAW,CAAC;AAAA,QACtC;AACA,iBAAS,KAAK,KAAK;AACnB,sBAAc,MAAM;AAAA,MACtB;AAEA,YAAM,WAAW,IAAI,WAAW,UAAU;AAC1C,UAAI,SAAS;AACb,iBAAW,SAAS,UAAU;AAC5B,iBAAS,IAAI,OAAO,MAAM;AAC1B,kBAAU,MAAM;AAAA,MAClB;AAEA,YAAM,aAAa,SAAS,SAAS;AACrC,YAAM,cAAc,aAAa,aAAa,GAAG,YAAY,UAAU;AACvE,YAAM,cAAc,YAAY,eAAe,CAAC;AAEhD,YAAM,WAAW,IAAI,SAAS,SAAS,MAAM;AAC7C,eAAS,IAAI,GAAG,IAAI,YAAY,KAAK;AACnC,cAAM,QAAQ,SAAS,SAAS,IAAI,GAAG,IAAI;AAC3C,oBAAY,CAAC,IAAI,QAAQ;AAAA,MAC3B;AAEA,YAAM,SAAS,aAAa,mBAAmB;AAC/C,aAAO,SAAS;AAChB,aAAO,QAAQ,aAAa,WAAW;AACvC,aAAO,UAAU,MAAM,aAAa;AACpC,aAAO,MAAM,CAAC;AACd,oBAAc;AAAA,IAChB,SAAS,KAAK;AACZ,gBAAU,GAAY;AAAA,IACxB;AAAA,EACF;AAEA,WAAS,YAAY;AACnB,QAAI,aAAa;AACf,kBAAY,KAAK;AACjB,oBAAc;AAAA,IAChB;AACA,QAAI,cAAc;AAChB,mBAAa,MAAM;AACnB,qBAAe;AAAA,IACjB;AAAA,EACF;AAEA,WAAS,QAAQ;AACf,QAAI,IAAI;AACN,SAAG,MAAM;AACT,WAAK;AAAA,IACP;AACA,cAAU;AAAA,EACZ;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA,MAAM;AAAA,IACN,MAAM;AAAA,EACR;AACF;AAEA,IAAO,qBAAQ,CACb,eAC6C;AAC7C,SAAO,CAAC,WAA4B;AAClC,WAAO,gBAAgB,EAAE,GAAG,YAAY,GAAG,OAAO,CAAC;AAAA,EACrD;AACF;","names":[]}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@amaster.ai/tts-client",
-  "version": "1.1.0-beta.7",
+  "version": "1.1.0-beta.71",
   "description": "Qwen TTS Realtime WebSocket client with audio playback",
   "type": "module",
   "main": "./dist/index.cjs",
@@ -33,13 +33,18 @@
     "registry": "https://registry.npmjs.org/"
   },
   "devDependencies": {
+    "jsdom": "^23.2.0",
     "tsup": "^8.3.5",
-    "typescript": "~5.7.2"
+    "typescript": "~5.7.2",
+    "vitest": "^1.6.1"
   },
   "scripts": {
     "build": "tsup",
     "dev": "tsup --watch",
     "clean": "rm -rf dist *.tsbuildinfo",
-    "type-check": "tsc --noEmit"
+    "type-check": "tsc --noEmit",
+    "test": "vitest run",
+    "test:watch": "vitest",
+    "test:ui": "vitest --ui"
   }
 }