Update app.py

app.py

@@ -3,21 +3,52 @@ from gtts import gTTS
 import tempfile
 from transformers import pipeline
 import os
+import logging
+from huggingface_hub import login
+
+# Set up logging for better error tracking
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
 # Configure Hugging Face API Token (Replace with your actual token)
 HUGGINGFACE_API_TOKEN = "your_huggingface_api_token"  # Replace with your Hugging Face API token
 
+# Validate API token and log in to Hugging Face Hub
+if HUGGINGFACE_API_TOKEN == "your_huggingface_api_token":
+    logger.error("Please replace 'your_huggingface_api_token' with a valid Hugging Face API token from https://huggingface.co/settings/tokens")
+    raise ValueError("Invalid Hugging Face API token. Please set a valid token.")
+try:
+    login(token=HUGGINGFACE_API_TOKEN)
+    logger.info("Successfully logged in to Hugging Face Hub")
+except Exception as e:
+    logger.error(f"Failed to log in to Hugging Face Hub: {str(e)}")
+    raise
+
 # Initialize Hugging Face Whisper model for speech-to-text (multilingual support)
-whisper_pipeline = pipeline("automatic-speech-recognition", 
-                           model="openai/whisper-small", 
-                           token=HUGGINGFACE_API_TOKEN,
-                           device=-1)  # Use CPU if GPU unavailable
+try:
+    whisper_pipeline = pipeline(
+        "automatic-speech-recognition",
+        model="openai/whisper-tiny",
+        token=HUGGINGFACE_API_TOKEN,
+        device=-1  # Use CPU if GPU unavailable
+    )
+    logger.info("Whisper model initialized successfully")
+except Exception as e:
+    logger.error(f"Failed to initialize Whisper model: {str(e)}")
+    raise
 
 # Initialize Hugging Face text generation model for chatbot
-text_generation_pipeline = pipeline("text2text-generation", 
-                                   model="facebook/bart-large-cnn", 
-                                   token=HUGGINGFACE_API_TOKEN,
-                                   device=-1)  # Use CPU if GPU unavailable
+try:
+    text_generation_pipeline = pipeline(
+        "text2text-generation",
+        model="google/flan-t5-small",
+        token=HUGGINGFACE_API_TOKEN,
+        device=-1  # Use CPU if GPU unavailable
+    )
+    logger.info("Text generation model initialized successfully")
+except Exception as e:
+    logger.error(f"Failed to initialize text generation model: {str(e)}")
+    raise
 
 # Define chatbot knowledge base
 base_info = """
@@ -89,9 +120,13 @@ context = [base_info, mental_health, medical_assistance, medicine_recommendation
 
 # Function to get AI response using text generation model
 def get_llm_response(message):
-    full_context = "\n".join(context) + f"\nUser: {message}\nMindCare:"
-    response = text_generation_pipeline(full_context, max_length=500, num_return_sequences=1)[0]['generated_text']
-    return response
+    try:
+        full_context = "\n".join(context) + f"\nUser: {message}\nMindCare:"
+        response = text_generation_pipeline(full_context, max_length=500, num_return_sequences=1)[0]['generated_text']
+        return response
+    except Exception as e:
+        logger.error(f"Error generating response: {str(e)}")
+        return f"Error generating response: {str(e)}"
 
 # Function to process voice input and convert to text
 def process_voice_input(audio_file, language="en"):
@@ -108,29 +143,34 @@ def process_voice_input(audio_file, language="en"):
                                         generate_kwargs={"language": language_map.get(language, "english")})
         return transcription["text"]
     except Exception as e:
+        logger.error(f"Error processing audio: {str(e)}")
         return f"Error processing audio: {str(e)}"
 
 # Define chatbot response function with voice and text input
 def bot(message=None, audio_file=None, language="en", history=None):
-    if audio_file:
-        # Process voice input if provided
-        message = process_voice_input(audio_file, language)
-    
-    if not message:
-        return "No input provided. Please type a message or upload an audio file.", None
-    
-    # Get response from text generation model
-    response = get_llm_response(message)
-    
-    # Convert the response to speech using gTTS
-    tts = gTTS(text=response, lang=language, slow=False)
-    
-    # Save the audio file in a temporary directory
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_audio:
-        audio_path = temp_audio.name
-        tts.save(audio_path)
-    
-    return response, audio_path
+    try:
+        if audio_file:
+            # Process voice input if provided
+            message = process_voice_input(audio_file, language)
+        
+        if not message:
+            return "No input provided. Please type a message or upload an audio file.", None
+        
+        # Get response from text generation model
+        response = get_llm_response(message)
+        
+        # Convert the response to speech using gTTS
+        tts = gTTS(text=response, lang=language, slow=False)
+        
+        # Save the audio file in a temporary directory
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as temp_audio:
+            audio_path = temp_audio.name
+            tts.save(audio_path)
+        
+        return response, audio_path
+    except Exception as e:
+        logger.error(f"Error in bot function: {str(e)}")
+        return f"Error: {str(e)}", None
 
 # Create Gradio interface with voice and text input
 demo = gr.Interface(
@@ -149,4 +189,9 @@ demo = gr.Interface(
 )
 
 # Launch the Gradio app
-demo.launch(debug=True, share=True)
+if __name__ == "__main__":
+    try:
+        demo.launch(debug=True, share=True)
+    except Exception as e:
+        logger.error(f"Failed to launch Gradio app: {str(e)}")
+        raise