Update app.py

app.py

@@ -13,7 +13,6 @@ from tenacity import retry, stop_after_attempt, wait_fixed
 @retry(stop=stop_after_attempt(3), wait=wait_fixed(2))
 def load_whisper_model():
     try:
-        # Whisper for speech-to-text (English-only)
         model = pipeline(
             "automatic-speech-recognition",
             model="openai/whisper-tiny.en",
@@ -29,7 +28,6 @@ def load_whisper_model():
 @retry(stop=stop_after_attempt(3), wait=wait_fixed(2))
 def load_symptom_model():
     try:
-        # Symptom-2-Disease for health analysis
         model = pipeline(
             "text-classification",
             model="abhirajeshbhai/symptom-2-disease-net",
@@ -40,10 +38,22 @@ def load_symptom_model():
         return model
     except Exception as e:
         print(f"Failed to load Symptom-2-Disease model: {str(e)}")
-        raise
+        # Fallback to a generic model
+        try:
+            model = pipeline(
+                "text-classification",
+                model="distilbert-base-uncased",
+                device=-1
+            )
+            print("Fallback to distilbert-base-uncased model.")
+            return model
+        except Exception as fallback_e:
+            print(f"Fallback model failed: {str(fallback_e)}")
+            raise
 
 whisper = None
 symptom_classifier = None
+is_fallback_model = False
 
 try:
     whisper = load_whisper_model()
@@ -53,7 +63,9 @@ except Exception as e:
 try:
     symptom_classifier = load_symptom_model()
 except Exception as e:
-    print(f"Symptom-2-Disease model initialization failed after retries: {str(e)}")
+    print(f"Symptom model initialization failed after retries: {str(e)}")
+    symptom_classifier = None
+    is_fallback_model = True  # Track if fallback model is used
 
 def compute_file_hash(file_path):
     """Compute MD5 hash of a file to check uniqueness."""
@@ -79,7 +91,7 @@ def transcribe_audio(audio_file):
         temp_wav = f"/tmp/{os.path.basename(audio_file)}.wav"
         sf.write(temp_wav, audio, sr)
         
-        # Transcribe with beam search for accuracy
+        # Transcribe with beam search
         with torch.no_grad():
             result = whisper(temp_wav, generate_kwargs={"num_beams": 5})
         transcription = result.get("text", "").strip()
@@ -113,6 +125,9 @@ def analyze_symptoms(text):
         if result and isinstance(result, list) and len(result) > 0:
             prediction = result[0]["label"]
             score = result[0]["score"]
+            if is_fallback_model:
+                print("Warning: Using fallback model (distilbert-base-uncased). Results may be less accurate.")
+                prediction = f"{prediction} (using fallback model)"
             print(f"Health Prediction: {prediction}, Score: {score:.4f}")
             return prediction, score
         return "No health condition predicted", 0.0
@@ -140,6 +155,13 @@ def analyze_voice(audio_file):
         if "Error transcribing" in transcription:
             return transcription
         
+        # Check for medication-related queries
+        if "medicine" in transcription.lower() or "treatment" in transcription.lower():
+            feedback = "Error: This tool does not provide medication or treatment advice. Please describe symptoms only (e.g., 'I have a fever')."
+            feedback += f"\n\n**Debug Info**: Transcription = '{transcription}', File Hash = {file_hash}"
+            feedback += "\n**Disclaimer**: This is not a diagnostic tool. Consult a healthcare provider for medical advice."
+            return feedback
+        
         # Analyze symptoms
         prediction, score = analyze_symptoms(transcription)
         if "Error analyzing" in prediction:
@@ -182,7 +204,7 @@ iface = gr.Interface(
     inputs=gr.Audio(type="filepath", label="Record or Upload Voice"),
     outputs=gr.Textbox(label="Health Assessment Feedback"),
     title="Health Voice Analyzer",
-    description="Record or upload a voice sample describing symptoms for preliminary health assessment. Supports English (transcription), with symptom analysis in English. Use clear audio (WAV, 16kHz) describing symptoms like 'I have a cough.'"
+    description="Record or upload a voice sample describing symptoms (e.g., 'I have a fever') for preliminary health assessment. Supports English only. Use clear audio (WAV, 16kHz). Do not ask for medication or treatment advice."
 )
 
 if __name__ == "__main__":