Update app.py

app.py

@@ -36,9 +36,11 @@ for message in st.session_state.messages:
 if (prompt := st.chat_input("Your message")) or len(audio):
     # If it's coming from the audio recorder transcribe the message with whisper.cpp
     if len(audio) > 0:
-        # Check if audio is a bytes object (raw audio data)
+        # Debugging: Check the type of the audio object
+        st.write(f"Audio Type: {type(audio)}")
+
+        # Handle the case where audio is in a byte format
         if isinstance(audio, bytes):
-            # Try to load audio from bytes using AudioSegment
             try:
                 # Convert the raw byte data to an AudioSegment instance
                 audio_segment = AudioSegment.from_file(io.BytesIO(audio), format="wav")
@@ -46,6 +48,12 @@ if (prompt := st.chat_input("Your message")) or len(audio):
             except Exception as e:
                 st.error(f"Error processing audio: {e}")
                 prompt = "Sorry, there was an error processing your audio."
+
+        # Handle the case where audio is an AudioSegment object
+        elif isinstance(audio, AudioSegment):
+            # Process it directly since it's already an AudioSegment
+            prompt = inference(audio)
+
         else:
             st.error("The audio data is not in the expected format.")
             prompt = "Sorry, the audio format is not correct."