Update app.py

app.py

@@ -350,28 +350,30 @@ def process_file(file):
 
 def process_question(question, history):
     if not rag.thread_id:
-        return "Please upload a document first.", history
+        return "Please upload a document first.", history, "", None
     try:
         response = rag.ask_question(question)
         history.append({"role": "user", "content": question})
         history.append({"role": "assistant", "content": response})
-        return "", history
+        return "", history, "", None
     except Exception as e:
-        return "", history + [{"role": "assistant", "content": f"Error: {str(e)}"}]
+        return "", history + [{"role": "assistant", "content": f"Error: {str(e)}"}], "", None
 
 def process_audio(audio_file, history):
     if audio_file is None:
-        return "Please record or upload an audio file.", history
+        return "Please record or upload an audio file.", history, "", None
     try:
         transcript = rag.transcribe_audio(audio_file)
         if not rag.thread_id:
-            return "Please upload a document first.", history
+            return "Please upload a document first.", history, "", None
         response = rag.ask_question(transcript)
         history.append({"role": "user", "content": f"🎤 {transcript}"})
         history.append({"role": "assistant", "content": response})
-        return "", history
+        return "", history, "", None
     except Exception as e:
-        return "", history + [{"role": "assistant", "content": f"Error: {str(e)}"}]
+        history.append({"role": "user", "content": f"🎤 [Error transcribing audio: {str(e)}]"})
+        history.append({"role": "assistant", "content": "It seems there was an error while transcribing audio due to a technical issue. If there's anything specific from the document or any other questions you have regarding the content, please let me know, and I can assist you with that information."})
+        return "", history, "", None
 
 def process_audio_base64(audio_base64, history):
     if not audio_base64:
@@ -417,33 +419,27 @@ with gr.Blocks(css=custom_css, title="Document Q&A System") as demo:
     audio_input = gr.Audio(type="filepath", label="Record or Upload Audio", elem_classes="gradio-audio", visible=False)
     audio_status = gr.Textbox(label="Audio Status", interactive=False, elem_classes="status-text", visible=False)
 
-    # Document upload section
+    # Unified card for upload, chat, and voice
     with gr.Row(elem_classes="centered-main"):
         with gr.Column():
             with gr.Group(elem_classes="compact-box"):
-                gr.Markdown("<div class='section-title'>1. Upload Document</div>")
+                gr.Markdown("<div class='section-title'>Document Q&A</div>")
                 file_input = gr.File(label="Upload Document", file_types=[".pdf", ".txt", ".doc", ".docx"], file_count="single", type="binary", elem_classes="upload-btn")
                 reset_btn = gr.Button("Reset Chat & Upload New Document", elem_classes="reset-btn")
+                file_input.change(process_file, file_input, file_output)
                 def reset_all():
                     rag.thread_id = None
-                    return "", [], "", None  # file_output, chatbot, question, audio_input
-                file_input.change(process_file, file_input, file_output)
+                    return "", [], "", None
                 reset_btn.click(reset_all, None, [file_output, chatbot, question, audio_input])
-
-    # Unified chat and voice panel
-    with gr.Row(elem_classes="centered-main"):
-        with gr.Column():
-            with gr.Group(elem_classes="compact-box"):
-                gr.Markdown("<div class='section-title'>2. Chat with your Document</div>")
+                file_output.render()
+                chatbot.render()
                 with gr.Row():
+                    question.render()
                     send_btn = gr.Button("Send", elem_classes="send-btn")
                     mic_btn = gr.Button("🎤", elem_classes="audio-btn")
-                question.submit(process_question, [question, chatbot], [question, chatbot, question, audio_input])
                 send_btn.click(process_question, [question, chatbot], [question, chatbot, question, audio_input])
-
-                # Inline audio recorder, hidden by default
+                question.submit(process_question, [question, chatbot], [question, chatbot, question, audio_input])
                 audio_input.change(process_audio, [audio_input, chatbot], [audio_input, chatbot, question, audio_input])
-
                 def show_audio():
                     return {audio_input: gr.update(visible=True), audio_status: gr.update(visible=True)}
                 mic_btn.click(show_audio, None, [audio_input, audio_status])