File size: 10,692 Bytes
f56a29b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 | import { useState, useRef, useCallback } from 'react';
import { ASR_PROVIDERS } from '@/lib/audio/constants';
import { createLogger } from '@/lib/logger';
const log = createLogger('AudioRecorder');
// TypeScript declarations for Web Speech API
declare global {
interface Window {
// eslint-disable-next-line @typescript-eslint/no-explicit-any -- Web Speech API not typed in lib.dom
SpeechRecognition: any;
// eslint-disable-next-line @typescript-eslint/no-explicit-any -- Web Speech API not typed in lib.dom
webkitSpeechRecognition: any;
}
}
export interface UseAudioRecorderOptions {
onTranscription?: (text: string) => void;
onError?: (error: string) => void;
}
export function useAudioRecorder(options: UseAudioRecorderOptions = {}) {
const { onTranscription, onError } = options;
const [isRecording, setIsRecording] = useState(false);
const [isProcessing, setIsProcessing] = useState(false);
const [recordingTime, setRecordingTime] = useState(0);
const mediaRecorderRef = useRef<MediaRecorder | null>(null);
const audioChunksRef = useRef<Blob[]>([]);
const timerRef = useRef<NodeJS.Timeout | null>(null);
// eslint-disable-next-line @typescript-eslint/no-explicit-any -- Web Speech API not typed
const speechRecognitionRef = useRef<any>(null);
// Synchronous lock to prevent rapid re-entry (React state updates are async)
const busyRef = useRef(false);
// Send audio to server for transcription
const transcribeAudio = useCallback(
async (audioBlob: Blob) => {
setIsProcessing(true);
try {
const formData = new FormData();
formData.append('audio', audioBlob, 'recording.webm');
// Get current ASR configuration from settings store
// Note: This requires importing useSettingsStore in browser context
if (typeof window !== 'undefined') {
const { useSettingsStore } = await import('@/lib/store/settings');
const { asrProviderId, asrLanguage, asrProvidersConfig } = useSettingsStore.getState();
formData.append('providerId', asrProviderId);
formData.append(
'modelId',
asrProvidersConfig?.[asrProviderId]?.modelId ||
ASR_PROVIDERS[asrProviderId as keyof typeof ASR_PROVIDERS]?.defaultModelId ||
'',
);
formData.append('language', asrLanguage);
// Append API key and base URL if configured
const providerConfig = asrProvidersConfig?.[asrProviderId];
if (providerConfig?.apiKey?.trim()) {
formData.append('apiKey', providerConfig.apiKey);
}
const effectiveBaseUrl =
providerConfig?.baseUrl?.trim() || providerConfig?.customDefaultBaseUrl || '';
if (effectiveBaseUrl) {
formData.append('baseUrl', effectiveBaseUrl);
}
}
const response = await fetch('/api/transcription', {
method: 'POST',
body: formData,
});
if (!response.ok) {
const error = await response.json();
throw new Error(error.error || 'Transcription failed');
}
const result = await response.json();
onTranscription?.(result.text);
} catch (error) {
log.error('Transcription error:', error);
onError?.(error instanceof Error ? error.message : '语音识别失败,请重试');
} finally {
setIsProcessing(false);
setRecordingTime(0);
}
},
[onTranscription, onError],
);
// Start recording
const startRecording = useCallback(async () => {
// Synchronous lock — React state is async so isRecording may be stale
if (busyRef.current) return;
busyRef.current = true;
try {
// Get current ASR configuration
if (typeof window !== 'undefined') {
const { useSettingsStore } = await import('@/lib/store/settings');
const { asrProviderId, asrLanguage } = useSettingsStore.getState();
// Use browser native ASR if configured
if (asrProviderId === 'browser-native') {
// Check if Speech Recognition is supported
if (!window.SpeechRecognition && !window.webkitSpeechRecognition) {
onError?.('您的浏览器不支持语音识别功能');
return;
}
const SpeechRecognition = window.SpeechRecognition || window.webkitSpeechRecognition;
const recognition = new SpeechRecognition();
recognition.lang = asrLanguage || 'zh-CN';
recognition.continuous = false;
recognition.interimResults = false;
recognition.onstart = () => {
setIsRecording(true);
setRecordingTime(0);
// Start timer
timerRef.current = setInterval(() => {
setRecordingTime((prev) => prev + 1);
}, 1000);
};
recognition.onresult = (event: {
results: {
[index: number]: { [index: number]: { transcript: string } };
};
}) => {
const transcript = event.results[0][0].transcript;
onTranscription?.(transcript);
};
recognition.onerror = (event: { error: string }) => {
log.error('Speech recognition error:', event.error);
let errorMessage = '语音识别失败';
switch (event.error) {
case 'aborted':
// Non-fatal: caused by our own cancel/stop logic or rapid toggle
busyRef.current = false;
setIsRecording(false);
setRecordingTime(0);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
return;
case 'no-speech':
errorMessage = '未检测到语音输入';
break;
case 'audio-capture':
errorMessage = '无法访问麦克风';
break;
case 'not-allowed':
errorMessage = '麦克风权限被拒绝';
break;
case 'network':
errorMessage = '网络错误';
break;
default:
errorMessage = `语音识别错误: ${event.error}`;
}
onError?.(errorMessage);
busyRef.current = false;
setIsRecording(false);
setRecordingTime(0);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
};
recognition.onend = () => {
busyRef.current = false;
setIsRecording(false);
setRecordingTime(0);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
};
recognition.start();
speechRecognitionRef.current = recognition;
return;
}
}
// Use MediaRecorder for server-side ASR
// Request microphone permission
const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
// Create MediaRecorder
const mediaRecorder = new MediaRecorder(stream, {
mimeType: 'audio/webm',
});
mediaRecorderRef.current = mediaRecorder;
audioChunksRef.current = [];
mediaRecorder.ondataavailable = (event) => {
if (event.data.size > 0) {
audioChunksRef.current.push(event.data);
}
};
mediaRecorder.onstop = async () => {
// Stop all audio tracks
stream.getTracks().forEach((track) => track.stop());
// Merge audio chunks
const audioBlob = new Blob(audioChunksRef.current, {
type: 'audio/webm',
});
// Send to server for transcription
await transcribeAudio(audioBlob);
busyRef.current = false;
};
// Start recording
mediaRecorder.start();
setIsRecording(true);
setRecordingTime(0);
// Start timer
timerRef.current = setInterval(() => {
setRecordingTime((prev) => prev + 1);
}, 1000);
} catch (error) {
busyRef.current = false;
log.error('Failed to start recording:', error);
onError?.('无法访问麦克风,请检查权限设置');
}
}, [onTranscription, onError, transcribeAudio]);
// Stop recording
const stopRecording = useCallback(() => {
// Stop Speech Recognition if active
if (speechRecognitionRef.current) {
speechRecognitionRef.current.stop();
speechRecognitionRef.current = null;
busyRef.current = false;
setIsRecording(false);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
return;
}
// Stop MediaRecorder if active
if (mediaRecorderRef.current && isRecording) {
mediaRecorderRef.current.stop();
busyRef.current = false;
setIsRecording(false);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
}
}, [isRecording]);
// Cancel recording
const cancelRecording = useCallback(() => {
// Cancel Speech Recognition if active
if (speechRecognitionRef.current) {
speechRecognitionRef.current.onresult = null; // Prevent transcription callback
speechRecognitionRef.current.onerror = null; // Suppress browser abort error events
speechRecognitionRef.current.stop();
speechRecognitionRef.current = null;
busyRef.current = false;
setIsRecording(false);
setRecordingTime(0);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
return;
}
// Cancel MediaRecorder if active
if (mediaRecorderRef.current && isRecording) {
// Stop recording without transcription
mediaRecorderRef.current.ondataavailable = null;
mediaRecorderRef.current.onstop = null;
mediaRecorderRef.current.stop();
// Stop all audio tracks
if (mediaRecorderRef.current.stream) {
mediaRecorderRef.current.stream.getTracks().forEach((track) => track.stop());
}
busyRef.current = false;
setIsRecording(false);
setRecordingTime(0);
if (timerRef.current) {
clearInterval(timerRef.current);
timerRef.current = null;
}
audioChunksRef.current = [];
}
}, [isRecording]);
return {
isRecording,
isProcessing,
recordingTime,
startRecording,
stopRecording,
cancelRecording,
};
}
|