react-native-speechkit

🎤 A powerful React Native library for real-time speech recognition, audio recording, and playback on iOS and Android.

✨ Features

• 🎙️ Real-time Speech Recognition (partial results)
• 📼 Automatic Audio Recording during recognition
• ▶️ Audio Playback of recorded files
• 📱 Cross-Platform: iOS & Android
• 🔒 TypeScript: Full type definitions
• ⚡ Event-Driven: Listen for results, errors, and recording events

🎬 Demo

📦 Installation

npm install react-native-speechkit
# or
yarn add react-native-speechkit

iOS Setup

1. Install CocoaPods:

   cd ios && pod install

2. Required permissions are already added to your Info.plist:
   • NSMicrophoneUsageDescription
   • NSSpeechRecognitionUsageDescription

Android Setup

• The RECORD_AUDIO permission is already in AndroidManifest.xml.
• For Android 6.0+ you may need to request runtime permissions:

import { PermissionsAndroid, Platform } from 'react-native';

async function requestMicrophonePermission() {
  if (Platform.OS === 'android') {
    const granted = await PermissionsAndroid.request(
      PermissionsAndroid.PERMISSIONS.RECORD_AUDIO,
      {
        title: 'Microphone Permission',
        message:
          'This app needs access to your microphone for speech recognition.',
        buttonNeutral: 'Ask Me Later',
        buttonNegative: 'Cancel',
        buttonPositive: 'OK',
      }
    );
    return granted === PermissionsAndroid.RESULTS.GRANTED;
  }
  return true;
}

🚀 Usage

import {
  startSpeechRecognition,
  stopSpeechRecognition,
  addSpeechResultListener,
  addSpeechErrorListener,
  addSpeechFinishedListener,
  playAudio,
} from 'react-native-speechkit';

// Start speech recognition (optionally pass fileURLString and autoStopAfter in ms)
await startSpeechRecognition();

// Listen for results (partial and final)
const resultSub = addSpeechResultListener(({ text, isFinal }) => {
  console.log('Transcribed:', text, 'Final:', isFinal);
});

// Listen for errors
const errorSub = addSpeechErrorListener(({ error }) => {
  console.error('Error:', error);
});

// Listen for finished event (final result and audio path)
const finishedSub = addSpeechFinishedListener(({ finalResult, audioLocalPath }) => {
  console.log('Final result:', finalResult, 'Audio file:', audioLocalPath);
  // Play the recorded audio
  if (audioLocalPath) {
    playAudio(audioLocalPath);
  }
});
### `playAudio(filePath: string): Promise<string>`

Play an audio file at the given path (local file or URL). Returns a promise that resolves when playback starts.

// Stop recognition
stopSpeechRecognition();

// Remove listeners when done
resultSub.remove();
errorSub.remove();
finishedSub.remove();

🧩 API Reference

startSpeechRecognition(fileURLString?: string | null, autoStopAfter?: number | null): Promise<string>

Start speech recognition and audio recording. Optionally specify a file path and auto-stop duration (ms).

stopSpeechRecognition(): void

Stop the current recognition session.

addSpeechResultListener(listener: (data: { text: string; isFinal: boolean }) => void)

Subscribe to speech recognition results (partial and final). Returns a subscription with .remove().

addSpeechErrorListener(listener: (data: { error: string }) => void)

Subscribe to errors. Returns a subscription with .remove().

addSpeechFinishedListener(listener: (data: { finalResult: string; audioLocalPath: string }) => void)

Subscribe to the finished event, which provides the final recognized text and the local audio file path. Returns a subscription with .remove().

📝 Example

import { useState, useEffect } from 'react';
import { View, Button, Text } from 'react-native';
import {
  startSpeechRecognition,
  stopSpeechRecognition,
  addSpeechResultListener,
  addSpeechErrorListener,
  addSpeechFinishedListener,
  playAudio,
} from 'react-native-speechkit';

export default function App() {
  const [isRecording, setIsRecording] = useState(false);
  const [text, setText] = useState('');
  const [audioPath, setAudioPath] = useState('');

  useEffect(() => {
    const resultSub = addSpeechResultListener(({ text, isFinal }) => {
      setText(text + (isFinal ? ' (final)' : ''));
    });
    const errorSub = addSpeechErrorListener(() => setIsRecording(false));
    const finishedSub = addSpeechFinishedListener(
      ({ finalResult, audioLocalPath }) => {
        setText(finalResult);
        setAudioPath(audioLocalPath);
        setIsRecording(false);
      }
    );
    return () => {
      resultSub.remove();
      errorSub.remove();
      finishedSub.remove();
    };
  }, []);

  return (
    <View>
      <Text>{text || 'No text yet'}</Text>
      <Text>{audioPath ? `Audio: ${audioPath}` : ''}</Text>
      <Button
        title={isRecording ? 'Stop' : 'Start'}
        onPress={
          isRecording
            ? () => {
                stopSpeechRecognition();
                setIsRecording(false);
              }
            : async () => {
                await startSpeechRecognition();
                setIsRecording(true);
              }
        }
      />
      {audioPath ? (
        <Button title="Play Audio" onPress={() => playAudio(audioPath)} />
      ) : null}
    </View>
  );
}

📱 Platform Notes

iOS

• Requires iOS 10.0+
• Uses Apple's Speech framework
• Supports partial results
• Requires user permission for microphone & speech recognition

Android

• Requires Android API 21+
• Uses Android's SpeechRecognizer
• Supports partial results
• Requires RECORD_AUDIO permission

�️ Contributing

Contributions are welcome! Please read the Contributing Guide and Code of Conduct.

📄 License

MIT © ChasonJia