Update app.py

app.py

@@ -11,18 +11,21 @@ client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 def respond(audio, history, system_message, max_tokens, temperature, top_p):
     # Convert audio to text
     recognizer = sr.Recognizer()
-    with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as temp_audio_file:
-        temp_audio_file.write(audio)
-        temp_audio_file_path = temp_audio_file.name
-
+    temp_audio_path = None
     try:
-        with sr.AudioFile(temp_audio_file_path) as source:
+        # Save the uploaded file to a temporary path
+        with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as temp_file:
+            temp_file.write(audio)
+            temp_audio_path = temp_file.name
+
+        with sr.AudioFile(temp_audio_path) as source:
             audio_data = recognizer.record(source)
             message = recognizer.recognize_google(audio_data)
     except Exception as e:
         return "Error in recognizing audio", None
-
-    os.remove(temp_audio_file_path)
+    finally:
+        if temp_audio_path:
+            os.remove(temp_audio_path)
 
     # Prepare messages for the model
     messages = [{"role": "system", "content": system_message}]
@@ -61,7 +64,7 @@ def respond(audio, history, system_message, max_tokens, temperature, top_p):
 demo = gr.Interface(
     fn=respond,
     inputs=[
-        gr.Audio(source="upload", type="bytes", label="Upload your audio"),
+        gr.Audio(type="bytes", label="Record your audio"),
         gr.State(value=[], label="History"),
         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
@@ -71,7 +74,8 @@ demo = gr.Interface(
     outputs=[
         gr.Textbox(label="Response"),
         gr.Audio(label="Response Audio", type="file")
-    ]
+    ],
+    live=True
 )
 
 if __name__ == "__main__":