> This is a page from the ElevenLabs documentation. For a complete page index, fetch https://elevenlabs.io/docs/llms.txt. For the full documentation in a single file, fetch https://elevenlabs.io/docs/llms-full.txt.

# React Native SDK

> **Info**
>
> 有关 ElevenAgents 工作原理的说明，请参阅 [ElevenAgents 概览](/docs/zh/eleven-agents/overview)。

## 安装

在 React Native 项目中安装此软件包及其依赖项。

```shell
npm install @elevenlabs/react-native @livekit/react-native @livekit/react-native-webrtc livekit-client
```

> **Tip**
>
> 可在[此处](https://github.com/elevenlabs/packages/tree/main/examples/react-native-expo)
> 查看使用此 SDK 和 Expo 的示例应用。

> **Tip**
>
> 正在从早期版本升级？运行 `npx skills add elevenlabs/packages`，为 AI 编程智能体安装
> `elevenlabs:sdk-migration` skill，自动处理导入变更、`ConversationProvider` 包装及 API 更新。

## 要求

* 配备 LiveKit 依赖项的 React Native
* 已为平台配置麦克风权限
* 兼容 Expo（仅限开发构建）
* React Native 0.79.0+，或在旧版本中将 Metro 配置为解析 `package.json` 的 `"exports"`

> **Warning**
>
> 此 SDK 专为 Expo 框架设计和构建。由于依赖 LiveKit 的 WebRTC 实现，
> 它需要开发构建，无法与 Expo Go 一起使用。

> **Warning**
>
> `@elevenlabs/react-native` 通过 `package.json` 的
> [`"exports"`](https://nodejs.org/api/packages.html#exports) 字段解析其 React Native 入口点。Metro 从 Metro 0.82.0 开始才默认解析
> `"exports"`；该版本随 [React Native 0.79.0](https://reactnative.dev/blog/2025/04/08/react-native-0.79) 发布（2025 年 4 月 8 日）。如果
> 使用的 React Native 版本早于 0.79，请在 `metro.config.js` 中设置 `resolver.unstable_enablePackageExports =
>   true`，以便正确解析此软件包：
>
> ```js
> const { getDefaultConfig } = require("metro-config");
>
> const config = getDefaultConfig(__dirname);
> config.resolver.unstable_enablePackageExports = true;
>
> module.exports = config;
> ```

## 设置

### Provider 设置

使用 `ConversationProvider` 包装应用，以启用 ElevenAgents 功能。

```tsx
import { ConversationProvider } from "@elevenlabs/react-native";
import React from "react";

function App() {
  return (
    <ConversationProvider>
      <YourAppComponents />
    </ConversationProvider>
  );
}
```

> **Note**
>
> `@elevenlabs/react-native` 从 `@elevenlabs/react` 重新导出 `ConversationProvider` 和所有 Hook。
> API 与 Web React SDK 完全相同——完整 API 参考请参阅 [React SDK 文档](/docs/zh/eleven-agents/libraries/react)。

## 使用

可通过 `@elevenlabs/react-native` 使用 React SDK 的所有 Hook：

* **`useConversation`** — 整合所有状态和方法的便捷 Hook
* **`useConversationControls`** — 操作方法（startSession、endSession 等）
* **`useConversationStatus`** — 连接状态
* **`useConversationInput`** — 静音状态
* **`useConversationMode`** — 说话／聆听状态
* **`useConversationFeedback`** — 反馈可用性和提交
* **`useConversationClientTool`** — 动态注册客户端工具
* **`useRawConversation`** — 原始对话实例

### 开始对话

```tsx
import { useConversationControls, useConversationStatus } from "@elevenlabs/react-native";
import React from "react";
import { View, Text, Button } from "react-native";

function ConversationComponent() {
  const { startSession, endSession } = useConversationControls();
  const { status } = useConversationStatus();

  const handleStart = async () => {
    await startSession({
      agentId: "agent_7101k5zvyjhmfg983brhmhkd98n6",
    });
  };

  return (
    <View>
      <Text>Status: {status}</Text>
      <Button
        title={status === "connected" ? "End" : "Start"}
        onPress={status === "connected" ? endSession : handleStart}
      />
    </View>
  );
}
```

有关私有智能体、身份验证、客户端工具、覆盖设置及其他所有功能，请参阅 [React SDK 文档](/docs/zh/eleven-agents/libraries/react)。

## 实现示例

以下是使用 ElevenLabs Agents SDK 的完整 React Native 组件示例：

```tsx
import { ConversationProvider, useConversation } from "@elevenlabs/react-native";
import React, { useState } from "react";
import { View, Text, TouchableOpacity, StyleSheet } from "react-native";

function ConversationScreen() {
  const [isConnected, setIsConnected] = useState(false);

  const conversation = useConversation({
    onConnect: () => {
      console.log("Connected to conversation");
      setIsConnected(true);
    },
    onDisconnect: () => {
      console.log("Disconnected from conversation");
      setIsConnected(false);
    },
    onMessage: (message) => {
      console.log("Message received:", message);
    },
    onError: (error) => {
      console.error("Conversation error:", error);
    },
  });

  const startConversation = async () => {
    try {
      await conversation.startSession({
        agentId: "agent_7101k5zvyjhmfg983brhmhkd98n6",
      });
    } catch (error) {
      console.error("Failed to start conversation:", error);
    }
  };

  const endConversation = async () => {
    try {
      await conversation.endSession();
    } catch (error) {
      console.error("Failed to end conversation:", error);
    }
  };

  return (
    <View style={styles.container}>
      <Text style={styles.status}>Status: {conversation.status}</Text>

      <Text style={styles.speaking}>
        Agent is {conversation.isSpeaking ? "speaking" : "not speaking"}
      </Text>

      <TouchableOpacity
        style={[styles.button, isConnected && styles.buttonActive]}
        onPress={isConnected ? endConversation : startConversation}
      >
        <Text style={styles.buttonText}>
          {isConnected ? "End Conversation" : "Start Conversation"}
        </Text>
      </TouchableOpacity>

      {conversation.canSendFeedback && (
        <View style={styles.feedbackContainer}>
          <TouchableOpacity
            style={styles.feedbackButton}
            onPress={() => conversation.sendFeedback(true)}
          >
            <Text>👍</Text>
          </TouchableOpacity>
          <TouchableOpacity
            style={styles.feedbackButton}
            onPress={() => conversation.sendFeedback(false)}
          >
            <Text>👎</Text>
          </TouchableOpacity>
        </View>
      )}
    </View>
  );
}

function App() {
  return (
    <ConversationProvider>
      <ConversationScreen />
    </ConversationProvider>
  );
}

const styles = StyleSheet.create({
  container: {
    flex: 1,
    justifyContent: "center",
    alignItems: "center",
    padding: 20,
  },
  status: {
    fontSize: 16,
    marginBottom: 10,
  },
  speaking: {
    fontSize: 14,
    marginBottom: 20,
    color: "#666",
  },
  button: {
    backgroundColor: "#007AFF",
    paddingHorizontal: 20,
    paddingVertical: 10,
    borderRadius: 8,
    marginBottom: 20,
  },
  buttonActive: {
    backgroundColor: "#FF3B30",
  },
  buttonText: {
    color: "white",
    fontSize: 16,
    fontWeight: "600",
  },
  feedbackContainer: {
    flexDirection: "row",
    gap: 10,
  },
  feedbackButton: {
    backgroundColor: "#F2F2F7",
    padding: 10,
    borderRadius: 8,
  },
});

export default App;
```

## 平台特定注意事项

### iOS

确保在 `Info.plist` 中正确配置麦克风权限：

```xml
<key>NSMicrophoneUsageDescription</key>
<string>This app needs microphone access to enable voice conversations with AI agents.</string>
```

### Android

在 `android/app/src/main/AndroidManifest.xml` 中添加麦克风权限：

```xml
<uses-permission android:name="android.permission.RECORD_AUDIO" />
```

建议在开始对话前请求运行时权限：

```tsx
import { PermissionsAndroid, Platform } from "react-native";

const requestMicrophonePermission = async () => {
  if (Platform.OS === "android") {
    const granted = await PermissionsAndroid.request(PermissionsAndroid.PERMISSIONS.RECORD_AUDIO, {
      title: "Microphone Permission",
      message: "This app needs microphone access to enable voice conversations.",
      buttonNeutral: "Ask Me Later",
      buttonNegative: "Cancel",
      buttonPositive: "OK",
    });
    return granted === PermissionsAndroid.RESULTS.GRANTED;
  }
  return true;
};
```