aiData/DouYin/transcribe_videos.py

import os
import logging
import sys
import time

# Ensure project root is in path
sys.path.append(r"d:\dsWork\aiData")
from Util.ASRClient import ASRClient

# Configure logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s', stream=sys.stdout)
logger = logging.getLogger("Transcriber")

def transcribe_all():
    audio_dir = r"d:\dsWork\aiData\DouYin\Audios"
    transcript_dir = r"d:\dsWork\aiData\DouYin\Transcripts"
    
    if not os.path.exists(transcript_dir):
        os.makedirs(transcript_dir)
        
    client = ASRClient()
    
    files = [f for f in os.listdir(audio_dir) if f.endswith(".mp3")]
    # Sort files to ensure deterministic order
    files.sort()
    logger.info(f"Found {len(files)} audio files.")
    
    for filename in files:
        audio_path = os.path.join(audio_dir, filename)
        name, _ = os.path.splitext(filename)
        txt_filename = f"{name}.txt"
        txt_path = os.path.join(transcript_dir, txt_filename)
        
        if os.path.exists(txt_path):
            # Check if file is empty
            if os.path.getsize(txt_path) > 0:
                logger.info(f"Skipping (already exists): {txt_filename}")
                continue
            else:
                logger.info(f"Re-processing empty file: {txt_filename}")
            
        logger.info(f"Processing: {filename}")
        
        try:
            # Direct local file transcription using synchronous method
            text = client.transcribe_file_sync(audio_path)
            
            if text:
                with open(txt_path, 'w', encoding='utf-8') as f:
                    f.write(text)
                logger.info(f"Saved transcript to: {txt_filename}")
            else:
                logger.error(f"Failed to transcribe: {filename}")
            
            # Add a small delay between files
            time.sleep(1)
                
        except Exception as e:
            logger.error(f"Error processing {filename}: {str(e)}", exc_info=True)

if __name__ == "__main__":
    try:
        transcribe_all()
    except KeyboardInterrupt:
        logger.info("Stopped by user")
    except Exception as e:
        logger.error(f"Fatal error: {str(e)}", exc_info=True)