Modifications

app.py

@@ -89,7 +89,8 @@ class AppState:
     stream: np.ndarray | None = None
     sampling_rate: int = 0
     pause_detected: bool = False
-    started_talking = False
+    started_talking: bool =  False
+    stopped: bool = False
 
 
 def determine_pause(audio: np.ndarray, sampling_rate: int, state: AppState) -> bool:
@@ -171,7 +172,7 @@ def process_audio(audio: tuple, state: AppState):
 
 
 def response(state: AppState):
-    if not state.pause_detected:
+    if not state.pause_detected and not state.started_talking:
         return None, AppState()
 
     for mp3_bytes in speaking(state.stream, state.sampling_rate):
@@ -181,9 +182,8 @@ def response(state: AppState):
 
 
 def start_recording_user(state: AppState):
-    if state.pause_detected and state.started_talking:
+    if not state.stopped:
         return gr.Audio(recording=True)
-    return None
 
 with gr.Blocks() as demo:
     with gr.Row():
@@ -213,7 +213,7 @@ with gr.Blocks() as demo:
         [input_audio]
     )
     cancel = gr.Button("Stop Conversation", variant="stop")
-    cancel.click(lambda: AppState(), None, [state], cancels=[respond, restart])
+    cancel.click(lambda: (AppState(stopped=True), gr.Audio(recording=False)), None, [state, input_audio], cancels=[respond, restart])
 
 
 demo.launch()