125 lines
4.4 KiB
Python
125 lines
4.4 KiB
Python
import os
|
||
import subprocess
|
||
import uuid
|
||
from functools import partial
|
||
from aiogram import types, Dispatcher, F
|
||
import ffmpeg
|
||
|
||
import logging
|
||
|
||
logging.basicConfig(
|
||
filename="/shared_storage/bot_log.txt",
|
||
level=logging.DEBUG,
|
||
format="%(asctime)s - %(levelname)s - %(message)s"
|
||
)
|
||
|
||
|
||
async def handle_voice_and_video(message: types.Message, redis_service, storage_path: str):
|
||
file_id = None
|
||
if message.content_type == types.ContentType.VOICE:
|
||
file_id = message.voice.file_id
|
||
elif message.content_type == types.ContentType.VIDEO_NOTE:
|
||
file_id = message.video_note.file_id
|
||
|
||
if not file_id:
|
||
return
|
||
|
||
if not os.path.exists(storage_path):
|
||
logging.error(f"❌ storage_path ({storage_path}) не существует!")
|
||
|
||
file = await message.bot.get_file(file_id)
|
||
file_path = file.file_path
|
||
|
||
file_uuid = str(uuid.uuid4())
|
||
original_filename = os.path.basename(file_path)
|
||
temp_filename = f"{file_uuid}_{original_filename}"
|
||
|
||
os.makedirs(storage_path, exist_ok=True)
|
||
|
||
temp_destination = os.path.join(storage_path, temp_filename)
|
||
|
||
await message.bot.download_file(file_path, temp_destination)
|
||
|
||
if not os.path.exists(temp_destination):
|
||
logging.error(f"❌ Файл {temp_destination} не был загружен!")
|
||
else:
|
||
logging.error(f"✅ Файл {temp_destination} был загружен!")
|
||
|
||
wav_filename = f"{file_uuid}.wav"
|
||
wav_destination = os.path.join(storage_path, wav_filename)
|
||
|
||
convert_to_wav(temp_destination, wav_destination)
|
||
|
||
if not os.path.exists(wav_destination):
|
||
logging.error(f"❌ WAV-файл {wav_destination} не был создан!")
|
||
else:
|
||
logging.error(f"✅ WAV-файл успешно создан: {wav_destination}")
|
||
|
||
os.remove(temp_destination)
|
||
|
||
task_data = {
|
||
"uuid": file_uuid,
|
||
"file_path": wav_destination,
|
||
"user_id": message.from_user.id,
|
||
"chat_id": message.chat.id,
|
||
"message_id": message.message_id
|
||
}
|
||
|
||
await redis_service.publish_task(task_data)
|
||
await message.reply("Your message has been received, converted to WAV, and queued for processing.\nWaiting for transcription...")
|
||
|
||
text = await redis_service.wait_for_text(
|
||
user_id=message.from_user.id,
|
||
chat_id=message.chat.id,
|
||
message_id=message.message_id
|
||
)
|
||
if text:
|
||
await message.reply(f"Transcription result:\n{text}")
|
||
else:
|
||
await message.reply("Sorry, transcription result was not received within the timeout.")
|
||
|
||
def convert_to_wav1(input_file: str, output_file: str):
|
||
"""
|
||
Конвертирует любой аудио/видеофайл в .wav с частотой 16kHz, 1 канал (моно).
|
||
"""
|
||
try:
|
||
ffmpeg.input(input_file).output(
|
||
output_file,
|
||
format="wav",
|
||
acodec="pcm_s16le",
|
||
ac=1,
|
||
ar="16000"
|
||
).run(overwrite_output=True)
|
||
except Exception as e:
|
||
print(f"Error converting {input_file} to WAV: {e}")
|
||
|
||
def convert_to_wav(input_file: str, output_file: str):
|
||
"""
|
||
Конвертирует любой аудиофайл в WAV с частотой 16kHz, 1 канал (моно).
|
||
Логирует ошибки FFmpeg.
|
||
"""
|
||
try:
|
||
logging.error(f"🔄 Конвертация: {input_file} -> {output_file}")
|
||
|
||
command = [
|
||
"ffmpeg", "-y", "-i", input_file,
|
||
"-ar", "16000", "-ac", "1", "-c:a", "pcm_s16le",
|
||
output_file
|
||
]
|
||
|
||
result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
|
||
|
||
if result.returncode != 0:
|
||
logging.error(f"❌ Ошибка FFmpeg:\n{result.stderr.decode()}")
|
||
else:
|
||
logging.error(f"✅ FFmpeg успешно создал файл {output_file}")
|
||
|
||
except Exception as e:
|
||
logging.error(f"❌ Исключение при конвертации {input_file}: {e}")
|
||
|
||
def register_audio_handlers(dp: Dispatcher, redis_service, storage_path: str):
|
||
# Оборачиваем callback для передачи дополнительных аргументов
|
||
handler_callback = partial(handle_voice_and_video, redis_service=redis_service, storage_path=storage_path)
|
||
# Регистрируем хэндлер с фильтром по content_type
|
||
dp.message.register(handler_callback, F.content_type.in_({types.ContentType.VOICE, types.ContentType.VIDEO_NOTE}))
|