Update app.py

app.py

@@ -298,42 +298,43 @@ def find_matching_filename(transcribed_filename: str, reference_map: Dict[str, s
     log_message(f"⚠️ No matching filename found for: {transcribed_filename}", "WARNING")
     return None
 
-def check_ffmpeg() -> bool:
-    """Check if ffmpeg is installed"""
-    try:
-        subprocess.run(["ffmpeg", "-version"], capture_output=True, check=True)
-        return True
-    except (FileNotFoundError, subprocess.CalledProcessError):
-        return False
-
 def transcribe_audio(wav_path: str) -> Optional[Dict[str, Any]]:
-    """Transcribe audio file using Whisper."""
+    """Transcribe audio file using Whisper from Transformers."""
     log_message(f"🎤 Transcribing audio file: {wav_path}", "INFO")
     
     try:
-        # Check ffmpeg first
-        if not check_ffmpeg():
-            log_message("❌ ffmpeg not found. Install with: pip install ffmpeg-python or use system package manager", "ERROR")
-            log_message("   Windows (Chocolatey): choco install ffmpeg", "ERROR")
-            log_message("   Windows (Scoop): scoop install ffmpeg", "ERROR")
-            log_message("   Linux (Ubuntu): sudo apt-get install ffmpeg", "ERROR")
-            log_message("   macOS (Homebrew): brew install ffmpeg", "ERROR")
-            return None
-        
-        import whisper
-        
-        # Load model
-        log_message(f"Loading Whisper {WHISPER_MODEL} model...", "INFO")
-        model = whisper.load_model(WHISPER_MODEL)
+        from transformers import pipeline
+        import librosa
+        
+        # Load audio with librosa
+        log_message(f"Loading audio file: {wav_path}", "INFO")
+        audio, sr = librosa.load(wav_path, sr=16000)
+        
+        # Initialize Whisper pipeline
+        log_message(f"Loading Whisper {WHISPER_MODEL} model from Transformers...", "INFO")
+        pipe = pipeline(
+            "automatic-speech-recognition",
+            model=f"openai/whisper-{WHISPER_MODEL}",
+            device=0 if __import__('torch').cuda.is_available() else -1  # GPU if available, else CPU
+        )
         
         # Transcribe
-        result = model.transcribe(wav_path)
+        log_message("Transcribing audio...", "INFO")
+        result = pipe(audio)
+        
+        # Format result to match openai-whisper format
+        formatted_result = {
+            "text": result["text"],
+            "segments": [{"text": result["text"]}]
+        }
         
         log_message(f"✅ Successfully transcribed: {wav_path}", "INFO")
-        return result
+        return formatted_result
         
-    except ImportError:
-        log_message("❌ Whisper not installed. Install with: pip install openai-whisper", "ERROR")
+    except ImportError as e:
+        missing_lib = str(e)
+        log_message(f"❌ Missing library. Install with: pip install transformers librosa torch torchaudio", "ERROR")
+        log_message(f"   Error: {missing_lib}", "ERROR")
         return None
     except Exception as e:
         log_message(f"❌ Failed to transcribe {wav_path}: {str(e)}", "ERROR")
@@ -680,17 +681,12 @@ async def startup_event():
     """Auto-start processing when server starts"""
     log_message("🚀 Server startup: Checking dependencies...", "INFO")
     
-    # Check ffmpeg
-    if not check_ffmpeg():
-        log_message("⚠️ WARNING: ffmpeg is not installed!", "WARNING")
-        log_message("   Whisper requires ffmpeg for audio processing", "WARNING")
-        log_message("   Install with one of the following:", "WARNING")
-        log_message("   - Windows (Chocolatey): choco install ffmpeg", "WARNING")
-        log_message("   - Windows (Scoop): scoop install ffmpeg", "WARNING")
-        log_message("   - Linux (Ubuntu): sudo apt-get install ffmpeg", "WARNING")
-        log_message("   - macOS (Homebrew): brew install ffmpeg", "WARNING")
-    else:
-        log_message("✅ ffmpeg found", "INFO")
+    try:
+        import transformers
+        log_message("✅ Transformers found", "INFO")
+    except ImportError:
+        log_message("⚠️ WARNING: Transformers not installed!", "WARNING")
+        log_message("   Install with: pip install transformers librosa torch torchaudio", "WARNING")
     
     log_message("🚀 Server startup: Auto-starting processing loop", "INFO")