> Discover all available pages from the documentation index: https://mastra.zisheng.pro/llms.txt

# voice.connect()

`connect()` 方法建立 WebSocket 或 WebRTC 连接，用于实时语音对语音通信。在使用 `send()` 或 `answer()` 等其他实时功能之前，必须调用此方法。

## 使用示例

```typescript
import { OpenAIRealtimeVoice } from '@mastra/voice-openai-realtime'
import Speaker from '@mastra/node-speaker'

const speaker = new Speaker({
  sampleRate: 24100, // Audio sample rate in Hz - standard for high-quality audio on MacBook Pro
  channels: 1, // Mono audio output (as opposed to stereo which would be 2)
  bitDepth: 16, // Bit depth for audio quality - CD quality standard (16-bit resolution)
})

// Initialize a real-time voice provider
const voice = new OpenAIRealtimeVoice({
  realtimeConfig: {
    model: 'gpt-5.1-realtime',
    apiKey: process.env.OPENAI_API_KEY,
    options: {
      sessionConfig: {
        turn_detection: {
          type: 'server_vad',
          threshold: 0.6,
          silence_duration_ms: 1200,
        },
      },
    },
  },
  speaker: 'alloy', // Default voice
})
// Connect to the real-time service
await voice.connect()
// Now you can use real-time features
voice.on('speaker', stream => {
  stream.pipe(speaker)
})
// With connection options
await voice.connect({
  timeout: 10000, // 10 seconds timeout
  reconnect: true,
})
```

## 参数

**options** (`Record<string, unknown>`): Provider 特定的连接选项

## 返回值

返回 `Promise<void>`，在连接成功建立后完成。

## Provider 特定选项

每个实时 Voice Provider 为 `connect()` 方法支持的选项可能不同：

### OpenAI Realtime

**options** (`Options`): 配置选项。

**options.timeout** (`number`): 连接超时时间（毫秒）

**options.reconnect** (`boolean`): 连接中断时是否自动重新连接

## 与 `CompositeVoice` 一起使用

使用 `CompositeVoice` 时，`connect()` 方法会委托给已配置的实时 Provider：

```typescript
import { CompositeVoice } from '@mastra/core/voice'
import { OpenAIRealtimeVoice } from '@mastra/voice-openai-realtime'
const realtimeVoice = new OpenAIRealtimeVoice()
const voice = new CompositeVoice({
  realtime: realtimeVoice,
})
// This will use the OpenAIRealtimeVoice provider
await voice.connect()
```

## 注意事项

- 只有支持语音对语音功能的实时 Voice Provider 才实现此方法
- 如果在不支持此功能的 Voice Provider 上调用此方法，它会记录警告并立即完成
- 使用 `send()` 或 `answer()` 等其他实时方法之前，必须先建立连接
- 使用完 Voice 实例后，请调用 `close()` 以正确清理资源
- 某些 Provider 可能会在连接中断时自动重新连接，具体取决于其实现
- 连接错误通常会作为异常抛出，应对其进行捕获和处理

## 相关方法

- [voice.send()](https://mastra.zisheng.pro/reference/voice/voice.send)：向 Voice Provider 发送音频数据
- [voice.answer()](https://mastra.zisheng.pro/reference/voice/voice.answer)：触发 Voice Provider 进行响应
- [voice.close()](https://mastra.zisheng.pro/reference/voice/voice.close)：断开与实时服务的连接
- [voice.on()](https://mastra.zisheng.pro/reference/voice/voice.on)：为语音事件注册事件监听器