File size: 10,482 Bytes
7226ab4 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 |
/**
* βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
* CLASSE: AudioProcessor
* βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
* Gerencia STT (Speech-to-Text), TTS (Text-to-Speech) e processamento de Γ‘udio
* IntegraΓ§Γ£o com Deepgram e Google TTS
* βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
*/
const axios = require('axios');
const fs = require('fs');
const path = require('path');
const ffmpeg = require('fluent-ffmpeg');
const googleTTS = require('google-tts-api');
const ConfigManager = require('./ConfigManager');
class AudioProcessor {
constructor(logger = null) {
this.config = ConfigManager.getInstance();
this.logger = logger || console;
this.tempFolder = this.config.TEMP_FOLDER;
this.sttCache = new Map();
this.ttsCache = new Map();
}
/**
* Gera nome de arquivo aleatΓ³rio
*/
generateRandomFilename(ext = '') {
return path.join(
this.tempFolder,
`${Date.now()}-${Math.random().toString(36).slice(2, 8)}${ext ? '.' + ext : ''}`
);
}
/**
* Limpa arquivo apΓ³s uso
*/
async cleanupFile(filePath) {
try {
if (!filePath || !fs.existsSync(filePath)) return;
return new Promise((resolve) => {
fs.unlink(filePath, (err) => {
if (err && err.code !== 'ENOENT') {
this.logger.warn(`β οΈ Erro ao limpar ${path.basename(filePath)}: ${err.code}`);
}
resolve();
});
});
} catch (e) {
this.logger.error('Erro ao limpar arquivo:', e.message);
}
}
/**
* STT usando Deepgram
* Transcreve Γ‘udio para texto
*/
async speechToText(audioBuffer, language = 'pt') {
try {
if (!this.config.DEEPGRAM_API_KEY) {
this.logger.warn('β οΈ Deepgram API Key nΓ£o configurada');
return {
sucesso: false,
texto: '[Audio recebido mas Deepgram nΓ£o configurado]',
erro: 'API_KEY_MISSING'
};
}
this.logger.info('π Iniciando STT (Deepgram)...');
// Converte OGG para MP3
const audioPath = this.generateRandomFilename('ogg');
const convertedPath = this.generateRandomFilename('mp3');
fs.writeFileSync(audioPath, audioBuffer);
// Converte para MP3
await new Promise((resolve, reject) => {
ffmpeg(audioPath)
.toFormat('mp3')
.audioCodec('libmp3lame')
.on('end', resolve)
.on('error', reject)
.save(convertedPath);
});
const convertedBuffer = fs.readFileSync(convertedPath);
// Chama Deepgram API
this.logger.info('π€ Enviando para Deepgram...');
const response = await axios.post(
this.config.DEEPGRAM_API_URL,
convertedBuffer,
{
headers: {
'Authorization': `Token ${this.config.DEEPGRAM_API_KEY}`,
'Content-Type': 'audio/mpeg'
},
params: {
model: this.config.DEEPGRAM_MODEL,
language: language || this.config.STT_LANGUAGE,
smart_format: true,
punctuate: true,
diarize: false,
numerals: true
},
timeout: 30000
}
);
let textoTranscrito = '';
if (response.data?.results?.channels?.[0]?.alternatives?.[0]?.transcript) {
textoTranscrito = response.data.results.channels[0].alternatives[0].transcript.trim();
}
if (!textoTranscrito || textoTranscrito.length < 2) {
textoTranscrito = '[NΓ£o consegui entender claramente]';
}
// Limpeza
await Promise.all([
this.cleanupFile(audioPath),
this.cleanupFile(convertedPath)
]);
this.logger.info(`π STT Completo: ${textoTranscrito.substring(0, 80)}...`);
return {
sucesso: true,
texto: textoTranscrito,
fonte: 'Deepgram STT',
confidence: response.data?.results?.channels?.[0]?.alternatives?.[0]?.confidence || 0
};
} catch (error) {
this.logger.error('β Erro STT:', error.message);
let errorCode = 'UNKNOWN';
if (error.response?.status === 401) {
errorCode = 'INVALID_API_KEY';
} else if (error.code === 'ECONNREFUSED') {
errorCode = 'CONNECTION_FAILED';
}
return {
sucesso: false,
texto: '[Recebi seu Γ‘udio mas houve um erro na transcriΓ§Γ£o]',
erro: errorCode,
mensagem: error.message
};
}
}
/**
* TTS usando Google TTS
* Converte texto para Γ‘udio
*/
async textToSpeech(text, language = 'pt') {
try {
if (!text || text.length === 0) {
return {
sucesso: false,
error: 'Texto vazio'
};
}
// Verifica cache
const cacheKey = `${text.substring(0, 50)}_${language}`;
if (this.ttsCache.has(cacheKey)) {
this.logger.debug('πΎ TTS from cache');
return this.ttsCache.get(cacheKey);
}
this.logger.info('π Iniciando TTS (Google)...');
// Trunca texto se necessΓ‘rio (Google TTS tem limite)
const maxChars = 500;
const textTruncated = text.substring(0, maxChars);
const audioUrl = googleTTS.getAudioUrl(textTruncated, {
lang: language || this.config.TTS_LANGUAGE,
slow: this.config.TTS_SLOW,
host: 'https://translate.google.com'
});
const outputPath = this.generateRandomFilename('mp3');
// Download do Γ‘udio
const response = await axios({
url: audioUrl,
method: 'GET',
responseType: 'arraybuffer',
timeout: 15000
});
const audioBuffer = Buffer.from(response.data);
if (audioBuffer.length === 0) {
throw new Error('Audio buffer vazio');
}
fs.writeFileSync(outputPath, audioBuffer);
const stats = fs.statSync(outputPath);
if (stats.size > this.config.MAX_AUDIO_SIZE_MB * 1024 * 1024) {
await this.cleanupFile(outputPath);
return {
sucesso: false,
error: 'Γudio TTS muito grande'
};
}
const finalBuffer = fs.readFileSync(outputPath);
await this.cleanupFile(outputPath);
const result = {
sucesso: true,
buffer: finalBuffer,
fonte: 'Google TTS',
size: finalBuffer.length
};
// Cache
this.ttsCache.set(cacheKey, result);
if (this.ttsCache.size > 50) {
const firstKey = this.ttsCache.keys().next().value;
this.ttsCache.delete(firstKey);
}
this.logger.info(`π TTS Completo: ${textTruncated.substring(0, 50)}... (${finalBuffer.length} bytes)`);
return result;
} catch (error) {
this.logger.error('β Erro TTS:', error.message);
return {
sucesso: false,
error: 'Erro ao gerar TTS: ' + error.message
};
}
}
/**
* Detecta se Γ© Γ‘udio animado (apenas tipo)
*/
detectAudioType(buffer) {
if (!buffer || buffer.length < 12) return 'unknown';
const header = buffer.slice(0, 4).toString('hex').toLowerCase();
// OGG Vorbis
if (header === '4f676753') return 'ogg';
// RIFF (WAV)
if (header === '52494646') return 'wav';
// MP3
if (header === '494433' || header === 'fffb') return 'mp3';
// FLAC
if (header === '664c6143') return 'flac';
// AAC
if (header === 'fff1' || header === 'fff9') return 'aac';
return 'unknown';
}
/**
* Aplica efeito de Γ‘udio (nightcore, slow, bass, etc)
*/
async applyAudioEffect(inputBuffer, effect = 'normal') {
try {
const inputPath = this.generateRandomFilename('mp3');
const outputPath = this.generateRandomFilename('mp3');
fs.writeFileSync(inputPath, inputBuffer);
let audioFilter = '';
let speed = 1;
let pitch = 0;
switch (effect.toLowerCase()) {
case 'nightcore':
speed = 1.5;
pitch = 8;
audioFilter = 'asetrate=44100*1.5,atempo=1/1.5';
break;
case 'slow':
speed = 0.7;
audioFilter = 'atempo=0.7';
break;
case 'fast':
speed = 1.3;
audioFilter = 'atempo=1.3';
break;
case 'bass':
audioFilter = 'bass=g=10';
break;
case 'treble':
audioFilter = 'treble=g=10';
break;
case 'echo':
audioFilter = 'aecho=0.8:0.9:1000:0.3';
break;
default:
// normal - no filter
break;
}
if (!audioFilter) {
// Sem efeito, copia direto
await this.cleanupFile(inputPath);
return { sucesso: true, buffer: inputBuffer };
}
await new Promise((resolve, reject) => {
let cmd = ffmpeg(inputPath);
if (audioFilter) {
cmd = cmd.audioFilter(audioFilter);
}
cmd
.outputOptions('-q:a 5')
.on('end', resolve)
.on('error', reject)
.save(outputPath);
});
const resultBuffer = fs.readFileSync(outputPath);
await Promise.all([
this.cleanupFile(inputPath),
this.cleanupFile(outputPath)
]);
return {
sucesso: true,
buffer: resultBuffer,
effect: effect,
size: resultBuffer.length
};
} catch (error) {
this.logger.error(`β Erro ao aplicar efeito ${effect}:`, error.message);
return {
sucesso: false,
error: error.message
};
}
}
/**
* Limpa cache de TTS
*/
clearCache() {
this.sttCache.clear();
this.ttsCache.clear();
this.logger.info('πΎ Caches de Γ‘udio limpos');
}
/**
* Retorna estatΓsticas
*/
getStats() {
return {
sttCacheSize: this.sttCache.size,
ttsCacheSize: this.ttsCache.size,
deepgramConfigured: !!this.config.DEEPGRAM_API_KEY,
sttEnabled: this.config.FEATURE_STT_ENABLED,
ttsEnabled: this.config.FEATURE_TTS_ENABLED
};
}
}
module.exports = AudioProcessor;
|