Serge ZaigraeffandClaude Opus 4.6
2138651474
- Fix race condition: pass return_timestamps as parameter instead of mutating shared state - Remove /local/transcriptions endpoint (path traversal vulnerability) - Add timeout=30 and URL scheme validation to prevent SSRF - Add temp file cleanup via try/finally in all route handlers - AsyncTaskManager: add threading.Lock and TTL cleanup for completed tasks - Change audio_rate to 16000 (matches Whisper's expected sample rate) - Extract filename from URL via Content-Disposition/path - Detect base64 audio format via python-magic - Fix response_size_bytes to use json.dumps instead of str() - Move scipy import to module level - Clamp temperature to [0.0, 1.0] - Deduplicate _load_model() code - Fix docstrings, README structure, typo, log level Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
86 lines
3.5 KiB
Python
86 lines
3.5 KiB
Python
"""
|
|
Модуль transcription_service.py содержит класс TranscriptionService,
|
|
который отвечает за транскрибацию аудиофайлов.
|
|
"""
|
|
|
|
import json
|
|
import os
|
|
import time
|
|
import traceback
|
|
from typing import Dict, Tuple
|
|
import logging
|
|
|
|
from ..audio.utils import get_audio_duration
|
|
from ..history import save_history
|
|
|
|
logger = logging.getLogger('app.transcription_service')
|
|
|
|
|
|
class TranscriptionService:
|
|
"""Сервис для транскрибации аудиофайлов."""
|
|
|
|
def __init__(self, transcriber, config: Dict):
|
|
self.transcriber = transcriber
|
|
self.config = config
|
|
|
|
def transcribe(self, file_path: str, filename: str, params: Dict = None) -> Tuple[Dict, int]:
|
|
"""
|
|
Транскрибирует аудиофайл по пути.
|
|
|
|
Args:
|
|
file_path: Путь к аудиофайлу.
|
|
filename: Имя файла (для логов и истории).
|
|
params: Дополнительные параметры для транскрибации.
|
|
|
|
Returns:
|
|
Кортеж (JSON-ответ, HTTP-код).
|
|
"""
|
|
params = params or {}
|
|
language = params.get('language', self.config.get('language', 'en'))
|
|
temperature = max(0.0, min(1.0, float(params.get('temperature', 0.0))))
|
|
prompt = params.get('prompt', '')
|
|
|
|
# Проверяем, запрошены ли временные метки
|
|
return_timestamps = params.get('return_timestamps', self.config.get('return_timestamps', False))
|
|
if isinstance(return_timestamps, str):
|
|
return_timestamps = return_timestamps.lower() in ('true', 't', 'yes', 'y', '1')
|
|
|
|
try:
|
|
# Определяем длительность аудиофайла
|
|
try:
|
|
duration = get_audio_duration(file_path)
|
|
except Exception as e:
|
|
logger.error(f"Ошибка при определении длительности файла: {e}")
|
|
return {"error": f"Не удалось определить длительность аудиофайла: {e}"}, 500
|
|
|
|
start_time = time.time()
|
|
result = self.transcriber.process_file(file_path, return_timestamps=return_timestamps)
|
|
processing_time = time.time() - start_time
|
|
|
|
# Формируем ответ
|
|
if return_timestamps:
|
|
response = {
|
|
"segments": result.get("segments", []),
|
|
"text": result.get("text", ""),
|
|
"processing_time": processing_time,
|
|
"response_size_bytes": len(json.dumps(result, ensure_ascii=False).encode('utf-8')),
|
|
"duration_seconds": duration,
|
|
"model": os.path.basename(self.config["model_path"])
|
|
}
|
|
else:
|
|
response = {
|
|
"text": result,
|
|
"processing_time": processing_time,
|
|
"response_size_bytes": len(json.dumps(result, ensure_ascii=False).encode('utf-8')),
|
|
"duration_seconds": duration,
|
|
"model": os.path.basename(self.config["model_path"])
|
|
}
|
|
|
|
save_history(response, filename, self.config)
|
|
return response, 200
|
|
|
|
except Exception as e:
|
|
logger.error(f"Ошибка при транскрибации файла '{filename}': {str(e)}")
|
|
logger.error(f"Traceback: {traceback.format_exc()}")
|
|
return {"error": str(e)}, 500
|