main.py

from flask import Flask, request, jsonify
import tempfile
import os
from werkzeug.utils import secure_filename
import logging
from datetime import datetime
from flask_cors import CORS

# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

app = Flask(__name__)
CORS(app)

# Configuration
app.config['MAX_CONTENT_LENGTH'] = 100 * 1024 * 1024  # 100MB max file size
ALLOWED_EXTENSIONS = {'wav', 'mp3', 'mp4', 'mpeg', 'mpga', 'm4a', 'webm', 'flac'}

# Initialize model variable
model = None
MODEL_SIZE = "base"  # Change this to your preferred model size

def load_whisper_model():
    """Load Whisper model with proper error handling"""
    global model
    try:
        # Try importing openai-whisper first
        import whisper
        logger.info(f"Loading Whisper model: {MODEL_SIZE}")
        model = whisper.load_model(MODEL_SIZE)
        logger.info("Whisper model loaded successfully")
        return True
    except ImportError:
        logger.error("OpenAI Whisper not installed. Install with: pip install openai-whisper")
        return False
    except AttributeError as e:
        logger.error(f"Whisper import error: {e}")
        logger.error("Make sure you have the correct whisper package installed:")
        logger.error("pip uninstall whisper")
        logger.error("pip install openai-whisper")
        return False
    except Exception as e:
        logger.error(f"Error loading Whisper model: {e}")
        return False

# Try to load the model at startup
model_loaded = load_whisper_model()

def allowed_file(filename):
    """Check if the file extension is allowed"""
    return '.' in filename and \
           filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS

def format_timestamp(seconds):
    """Convert seconds to HH:MM:SS.mmm format"""
    hours = int(seconds // 3600)
    minutes = int((seconds % 3600) // 60)
    secs = seconds % 60
    return f"{hours:02d}:{minutes:02d}:{secs:06.3f}"

@app.route('/', methods=['GET'])
def health_check():
    """Health check endpoint"""
    return jsonify({
        "status": "healthy" if model_loaded else "unhealthy",
        "message": "Whisper Transcription API is running" if model_loaded else "Whisper model failed to load",
        "model": MODEL_SIZE if model_loaded else "none",
        "model_loaded": model_loaded,
        "timestamp": datetime.now().isoformat()
    })

@app.route('/transcribe', methods=['POST'])
def transcribe_audio():
    """
    Transcribe audio file and return word-level timestamps
    
    Expected form data:
    - audio_file: The audio file to transcribe
    - language (optional): Language code (e.g., 'en', 'es', 'fr')
    - task (optional): 'transcribe' or 'translate' (default: transcribe)
    """
    try:
        # Check if model is loaded
        if not model_loaded or model is None:
            return jsonify({
                'error': 'Whisper model not loaded. Please check server logs and ensure openai-whisper is installed correctly.'
            }), 503
        
        # Check if audio file is present
        if 'audio_file' not in request.files:
            return jsonify({'error': 'No audio file provided'}), 400
        
        file = request.files['audio_file']
        
        if file.filename == '':
            return jsonify({'error': 'No file selected'}), 400
        
        if not allowed_file(file.filename):
            return jsonify({
                'error': f'File type not allowed. Supported formats: {", ".join(ALLOWED_EXTENSIONS)}'
            }), 400
        
        # Get optional parameters
        language = request.form.get('language', None)
        task = request.form.get('task', 'transcribe')
        
        if task not in ['transcribe', 'translate']:
            return jsonify({'error': 'Task must be either "transcribe" or "translate"'}), 400
        
        # Save uploaded file temporarily
        with tempfile.NamedTemporaryFile(delete=False, suffix=f".{file.filename.rsplit('.', 1)[1].lower()}") as tmp_file:
            file.save(tmp_file.name)
            temp_path = tmp_file.name
        
        logger.info(f"Processing file: {file.filename}")
        
        try:
            # Transcribe with word-level timestamps
            result = model.transcribe(
                temp_path,
                language=language,
                task=task,
                word_timestamps=True,
                verbose=False
            )
            
            # Extract word-level data
            word_segments = []
            for segment in result.get("segments", []):
                if "words" in segment:
                    for word_data in segment["words"]:
                        word_segments.append({
                            "word": word_data.get("word", "").strip(),
                            "start": word_data.get("start", 0),
                            "end": word_data.get("end", 0),
                            "start_formatted": format_timestamp(word_data.get("start", 0)),
                            "end_formatted": format_timestamp(word_data.get("end", 0)),
                            "confidence": word_data.get("probability", 0)
                        })
            
            # Prepare response
            response_data = {
                "success": True,
                "filename": secure_filename(file.filename),
                "language": result.get("language", "unknown"),
                "task": task,
                "duration": result.get("segments", [{}])[-1].get("end", 0) if result.get("segments") else 0,
                "text": result.get("text", ""),
                "word_count": len(word_segments),
                "segments": result.get("segments", []),
                "words": word_segments,
                "model_used": MODEL_SIZE,
                "processing_time": None  # You can add timing if needed
            }
            
            logger.info(f"Successfully transcribed {len(word_segments)} words from {file.filename}")
            return jsonify(response_data)
            
        except Exception as e:
            logger.error(f"Transcription error: {str(e)}")
            return jsonify({'error': f'Transcription failed: {str(e)}'}), 500
            
        finally:
            # Clean up temporary file
            if os.path.exists(temp_path):
                os.unlink(temp_path)
    
    except Exception as e:
        logger.error(f"API error: {str(e)}")
        return jsonify({'error': f'Server error: {str(e)}'}), 500

@app.route('/models', methods=['GET'])
def available_models():
    """Get information about available Whisper models"""
    models_info = {
        "current_model": MODEL_SIZE if model_loaded else "none",
        "model_loaded": model_loaded,
        "available_models": {
            "tiny": {"size": "~39 MB", "speed": "~32x", "accuracy": "lowest"},
            "base": {"size": "~74 MB", "speed": "~16x", "accuracy": "low"},
            "small": {"size": "~244 MB", "speed": "~6x", "accuracy": "medium"},
            "medium": {"size": "~769 MB", "speed": "~2x", "accuracy": "high"},
            "large": {"size": "~1550 MB", "speed": "~1x", "accuracy": "highest"}
        },
        "supported_languages": [
            "en", "zh", "de", "es", "ru", "ko", "fr", "ja", "pt", "tr", "pl", "ca", "nl", 
            "ar", "sv", "it", "id", "hi", "fi", "vi", "he", "uk", "el", "ms", "cs", "ro", 
            "da", "hu", "ta", "no", "th", "ur", "hr", "bg", "lt", "la", "mi", "ml", "cy", 
            "sk", "te", "fa", "lv", "bn", "sr", "az", "sl", "kn", "et", "mk", "br", "eu", 
            "is", "hy", "ne", "mn", "bs", "kk", "sq", "sw", "gl", "mr", "pa", "si", "km", 
            "sn", "yo", "so", "af", "oc", "ka", "be", "tg", "sd", "gu", "am", "yi", "lo", 
            "uz", "fo", "ht", "ps", "tk", "nn", "mt", "sa", "lb", "my", "bo", "tl", "mg", 
            "as", "tt", "haw", "ln", "ha", "ba", "jw", "su"
        ],
        "installation_help": {
            "error": "Whisper model not loaded" if not model_loaded else None,
            "install_command": "pip install openai-whisper torch torchaudio",
            "uninstall_conflicts": "pip uninstall whisper (if you have conflicting whisper package)"
        }
    }
    return jsonify(models_info)

@app.errorhandler(413)
def too_large(e):
    return jsonify({'error': 'File too large. Maximum size is 100MB'}), 413

@app.errorhandler(404)
def not_found(e):
    return jsonify({'error': 'Endpoint not found'}), 404

@app.errorhandler(500)
def internal_error(e):
    return jsonify({'error': 'Internal server error'}), 500

if __name__ == '__main__':
    if not model_loaded:
        print(f"""
        ⚠️  WHISPER MODEL LOADING FAILED ⚠️
        ===================================
        
        The Whisper model could not be loaded. Please check:
        
        1. Install the correct package:
           pip install openai-whisper torch torchaudio
        
        2. If you have conflicts, uninstall the wrong whisper package:
           pip uninstall whisper
           pip install openai-whisper
        
        3. Make sure you have sufficient disk space for the model
        
        The server will start but transcription will not work until the model is loaded.
        """)
    else:
        print(f"""
        Whisper Transcription API Server
        ================================
        Model: {MODEL_SIZE} ✅
        Status: Ready
        Endpoints:
        - GET  /           : Health check
        - POST /transcribe : Transcribe audio file
        - GET  /models     : Available models info
        
        Supported formats: {', '.join(ALLOWED_EXTENSIONS)}
        Max file size: 100MB
        """)
    
    app.run(debug=True, host='0.0.0.0', port=7860)