Update app.py

app.py

@@ -44,45 +44,17 @@ model = SALMONN(
 model.to(args.device)
 model.eval()
 
-# gradio 
-def gradio_reset(chat_state):
+def gradio_answer(speech, text_input, num_beams, temperature, top_p):
     
-    chat_state = []
-    return (None,
-            gr.update(value=None, interactive=True),
-            gr.update(placeholder='Please upload your wav first', interactive=False),
-            gr.update(value="Upload & Start Chat", interactive=True),
-            chat_state)
-
-def upload_speech(gr_speech, text_input, chat_state):
-    
-    if gr_speech is None:
-        return None, None, gr.update(interactive=True), chat_state, None
-    chat_state.append(gr_speech)
-    return (gr.update(interactive=False),
-            gr.update(interactive=True, placeholder='Type and press Enter'),
-            gr.update(value="Start Chatting", interactive=False),
-            chat_state)
-
-def gradio_ask(user_message, chatbot, chat_state):
-    
-    if len(user_message) == 0:
-        return gr.update(interactive=True, placeholder='Input should not be empty!'), chatbot, chat_state
-    chat_state.append(user_message)
-    chatbot.append([user_message, None])
-    # 
-    return gr.update(interactive=False, placeholder='Currently only single round conversations are supported.'), chatbot, chat_state
-
-def gradio_answer(chatbot, chat_state, num_beams, temperature, top_p):
     llm_message = model.generate(
-        wav_path=chat_state[0],
-        prompt=chat_state[1],
+        wav_path=speech,
+        prompt=text_input,
         num_beams=num_beams,
         temperature=temperature,
         top_p=top_p,
     )
-    chatbot[-1][1] = llm_message[0]
-    return chatbot, chat_state
+    
+    return llm_message
 
 title = """<h1 align="center">SALMONN: Speech Audio Language Music Open Neural Network</h1>"""
 image_src = """<h1 align="center"><a href="https://github.com/bytedance/SALMONN"><img src="https://raw.githubusercontent.com/bytedance/SALMONN/main/resource/salmon.png", alt="SALMONN" border="0" style="margin: 0 auto; height: 200px;" /></a> </h1>"""
@@ -97,9 +69,7 @@ with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column():
             speech = gr.Audio(label="Audio", type='filepath')
-            upload_button = gr.Button(value="Upload & Start Chat", interactive=True, variant="primary")
-            clear = gr.Button("Restart")
-
+            
             num_beams = gr.Slider(
                 minimum=1,
                 maximum=10,
@@ -128,10 +98,9 @@ with gr.Blocks() as demo:
             )
 
         with gr.Column():
-            chat_state = gr.State([])
             
-            chatbot = gr.Chatbot(label='SALMONN-7B')
-            text_input = gr.Textbox(label='User', placeholder='Please upload your audio first', interactive=False)
+            text_input = gr.Textbox(label='User', placeholder='Please upload your audio first', interactive=True)
+            answer = gr.Textbox(label="Salmonn answer")
 
     with gr.Row():
         examples = gr.Examples(
@@ -157,12 +126,11 @@ with gr.Blocks() as demo:
             inputs=[speech, text_input]
         )
         
-    upload_button.click(upload_speech, [speech, text_input, chat_state], [speech, text_input, upload_button, chat_state])
-
-    text_input.submit(gradio_ask, [text_input, chatbot, chat_state], [text_input, chatbot, chat_state]).then(
-        gradio_answer, [chatbot, chat_state, num_beams, temperature, top_p], [chatbot, chat_state]
+    
+    text_input.submit(
+        gradio_answer, [speech, text_input, num_beams, temperature, top_p], [answer]
     )
-    clear.click(gradio_reset, [chat_state], [chatbot, speech, text_input, upload_button, chat_state], queue=False)
+    
 
 # demo.launch(share=True, enable_queue=True, server_port=int(args.port))
-demo.launch(share=True)
+demo.launch(share=False)