Update app.py

app.py

@@ -42,13 +42,13 @@ if st.sidebar.button("🔄 Clear Chat"):
 show_image = st.sidebar.checkbox("📖 Show Document Image", value=True)
 
 # ------------------ Split Layout ------------------
-col1, col2 = st.columns([1, 2])  # Adjust ratio as needed
+col1, col2 = st.columns([1, 2])
 
 # ------------------ Image Panel (Left) ------------------
 with col1:
     if show_image and st.session_state.image_url:
         st.image(st.session_state.image_url, caption="📑 Extracted Page", use_container_width=True)
-        st.session_state.image_updated = False  # Reset flag after rendering
+        st.session_state.image_updated = False
 
 # ------------------ Chat Panel (Right) ------------------
 with col2:
@@ -56,43 +56,46 @@ with col2:
         role, content = message["role"], message["content"]
         st.chat_message(role).write(content)
 
+    # --- Voice-to-Text using file upload (placeholder until OpenAI real-time integration) ---
+    st.markdown("### 🎙️ Upload a voice file to transcribe")
+    audio_file = st.file_uploader("Upload a .wav file", type=["wav"])
+
+    if audio_file is not None:
+        try:
+            transcript_response = client.audio.transcriptions.create(
+                model="whisper-1",
+                file=audio_file,
+                response_format="text"
+            )
+            transcript = transcript_response.strip()
+            st.success(f"📝 Transcription: {transcript}")
+            if st.button("✅ Send to Assistant"):
+                st.session_state.messages.append({"role": "user", "content": transcript})
+                st.rerun()
+        except Exception as e:
+            st.error(f"❌ Transcription failed: {str(e)}")
+
+    # --- Traditional text input ---
     if prompt := st.chat_input("Type your question about the document..."):
         st.session_state.messages.append({"role": "user", "content": prompt})
         st.chat_message("user").write(prompt)
 
         try:
-            # Initialize thread if needed
             if st.session_state.thread_id is None:
                 thread = client.beta.threads.create()
                 st.session_state.thread_id = thread.id
 
             thread_id = st.session_state.thread_id
+            client.beta.threads.messages.create(thread_id=thread_id, role="user", content=prompt)
+            run = client.beta.threads.runs.create(thread_id=thread_id, assistant_id=ASSISTANT_ID)
 
-            # Send message to assistant
-            client.beta.threads.messages.create(
-                thread_id=thread_id,
-                role="user",
-                content=prompt
-            )
-
-            # Run assistant
-            run = client.beta.threads.runs.create(
-                thread_id=thread_id,
-                assistant_id=ASSISTANT_ID
-            )
-
-            # Wait for assistant response
             with st.spinner("Assistant is thinking..."):
                 while True:
-                    run_status = client.beta.threads.runs.retrieve(
-                        thread_id=thread_id,
-                        run_id=run.id
-                    )
+                    run_status = client.beta.threads.runs.retrieve(thread_id=thread_id, run_id=run.id)
                     if run_status.status == "completed":
                         break
                     time.sleep(1)
 
-            # Get assistant response
             messages = client.beta.threads.messages.list(thread_id=thread_id)
             assistant_message = None
             for message in reversed(messages.data):
@@ -103,7 +106,6 @@ with col2:
             st.chat_message("assistant").write(assistant_message)
             st.session_state.messages.append({"role": "assistant", "content": assistant_message})
 
-            # Extract GitHub image from response if available
             image_match = re.search(
                 r'https://raw\.githubusercontent\.com/AndrewLORTech/surgical-pathology-manual/main/[\w\-/]*\.png',
                 assistant_message
@@ -111,7 +113,7 @@ with col2:
             if image_match:
                 st.session_state.image_url = image_match.group(0)
                 st.session_state.image_updated = True
-                st.rerun()  # Trigger rerun to refresh image display
+                st.rerun()
 
         except Exception as e:
             st.error(f"❌ Error: {str(e)}")