Update app.py

app.py

@@ -140,38 +140,43 @@ def real_time_analysis():
     st.info("Listening... Say 'stop' to end the process.")
 
     def audio_frame_callback(audio_frame):
-    # Convert audio frame to bytes
-    audio_data = audio_frame.to_ndarray()
-    audio_bytes = (audio_data * 32767).astype(np.int16).tobytes()  # Convert to int16 format
+        # Convert audio frame to bytes
+        audio_data = audio_frame.to_ndarray()
+        print(f"Audio data shape: {audio_data.shape}")  # Debug: Check audio data shape
+        print(f"Audio data sample: {audio_data[:10]}")  # Debug: Check first 10 samples
 
-    # Transcribe the audio
-    text = transcribe_audio(audio_bytes)
-    if text:
-        st.write(f"*Recognized Text:* {text}")
+        audio_bytes = (audio_data * 32767).astype(np.int16).tobytes()  # Convert to int16 format
 
-        # Analyze sentiment
-        sentiment, score = analyze_sentiment(text)
-        st.write(f"*Sentiment:* {sentiment} (Score: {score})")
+        # Transcribe the audio
+        text = transcribe_audio(audio_bytes)
+        if text:
+            st.write(f"*Recognized Text:* {text}")
 
-        # Handle objection
-        objection_response = handle_objection(text)
-        st.write(f"*Objection Response:* {objection_response}")
+            # Analyze sentiment
+            sentiment, score = analyze_sentiment(text)
+            st.write(f"*Sentiment:* {sentiment} (Score: {score})")
 
-        # Get product recommendation
-        recommendations = []
-        if is_valid_input(text) and is_relevant_sentiment(score):
-            query_embedding = model.encode([text])
-            distances, indices = product_recommender.index.search(query_embedding, 1)
+            # Handle objection
+            objection_response = handle_objection(text)
+            st.write(f"*Objection Response:* {objection_response}")
 
-            if distances[0][0] < 1.5:  # Similarity threshold
-                recommendations = product_recommender.get_recommendations(text)
+            # Get product recommendation
+            recommendations = []
+            if is_valid_input(text) and is_relevant_sentiment(score):
+                query_embedding = model.encode([text])
+                distances, indices = product_recommender.index.search(query_embedding, 1)
 
-        if recommendations:
-            st.write("*Product Recommendations:*")
-            for rec in recommendations:
-                st.write(rec)
+                if distances[0][0] < 1.5:  # Similarity threshold
+                    recommendations = product_recommender.get_recommendations(text)
 
-    return audio_frame
+            if recommendations:
+                st.write("*Product Recommendations:*")
+                for rec in recommendations:
+                    st.write(rec)
+        else:
+            st.error("No transcription returned.")  # Debug: Check if transcription fails
+
+        return audio_frame
 
     # Start WebRTC audio stream
     webrtc_ctx = webrtc_streamer(
@@ -272,5 +277,4 @@ def run_app():
             st.error(f"Error loading dashboard: {e}")
 
 if __name__ == "__main__":
-    run_app()
-    
+    run_app()