Update app.py

app.py

@@ -1,12 +1,8 @@
 import streamlit as st
-import numpy as np
-import tempfile
-import soundfile as sf
 import os
 import time
 import re
 from openai import OpenAI
-from streamlit_audio_recorder import audio_recorder
 
 # ------------------ App Configuration ------------------
 st.set_page_config(page_title="Document AI Assistant", layout="wide")
@@ -19,7 +15,7 @@ ASSISTANT_ID = os.environ.get("ASSISTANT_ID")
 
 # ------------------ Error Handling for Missing Secrets ------------------
 if not OPENAI_API_KEY or not ASSISTANT_ID:
-    st.error("❌ Missing secrets. Please ensure both OPENAI_API_KEY and ASSISTANT_ID are set in your Hugging Face Space secrets.")
+    st.error("Missing secrets. Please ensure both OPENAI_API_KEY and ASSISTANT_ID are set in your Hugging Face Space secrets.")
     st.stop()
 
 client = OpenAI(api_key=OPENAI_API_KEY)
@@ -33,8 +29,6 @@ if "image_url" not in st.session_state:
     st.session_state.image_url = None
 if "image_updated" not in st.session_state:
     st.session_state.image_updated = False
-if "transcript_preview" not in st.session_state:
-    st.session_state.transcript_preview = None
 
 # ------------------ Sidebar Controls ------------------
 st.sidebar.header("🔧 Settings")
@@ -43,87 +37,62 @@ if st.sidebar.button("🔄 Clear Chat"):
     st.session_state.thread_id = None
     st.session_state.image_url = None
     st.session_state.image_updated = False
-    st.session_state.transcript_preview = None
     st.rerun()
 
 show_image = st.sidebar.checkbox("📖 Show Document Image", value=True)
-preview_toggle = st.sidebar.checkbox("📝 Preview transcription before sending", value=True)
 
 # ------------------ Split Layout ------------------
-col1, col2 = st.columns([1, 2])
+col1, col2 = st.columns([1, 2])  # Adjust ratio as needed
 
 # ------------------ 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
-
-# ------------------ Voice Input Processing ------------------
-with col2:
-    st.markdown("### 🎙️ Voice Input (Optional)")
-    audio_bytes = audio_recorder(pause_threshold=2.0)
-
-    if audio_bytes:
-        st.info("Transcribing your voice...")
-        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp:
-            tmp.write(audio_bytes)
-            tmp_path = tmp.name
-
-        with open(tmp_path, "rb") as audio_file:
-            try:
-                whisper_result = client.audio.transcriptions.create(
-                    model="whisper-1",
-                    file=audio_file,
-                    response_format="json"
-                )
-                transcript = whisper_result.text.strip()
-                confidence = whisper_result.get("confidence", "N/A")
-
-                if transcript:
-                    st.success(f"Recognized: {transcript}")
-                    st.caption(f"🧠 Confidence: {confidence}")
-                    if preview_toggle:
-                        st.session_state.transcript_preview = transcript
-                    else:
-                        st.session_state.messages.append({"role": "user", "content": transcript})
-                        st.rerun()
-            except Exception as e:
-                st.error(f"❌ Transcription failed: {str(e)}")
-
-    if st.session_state.transcript_preview:
-        st.markdown("---")
-        st.markdown("### 📝 Transcription Preview")
-        st.markdown(f"> {st.session_state.transcript_preview}")
-        if st.button("✅ Send to Assistant"):
-            st.session_state.messages.append({"role": "user", "content": st.session_state.transcript_preview})
-            st.session_state.transcript_preview = None
-            st.rerun()
-        if st.button("❌ Discard"):
-            st.session_state.transcript_preview = None
-            st.rerun()
+        st.session_state.image_updated = False  # Reset flag after rendering
 
 # ------------------ Chat Panel (Right) ------------------
 with col2:
+    for message in st.session_state.messages:
+        role, content = message["role"], message["content"]
+        st.chat_message(role).write(content)
+
     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)
 
-            with st.spinner("🤖 Assistant is thinking..."):
+            # 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):
@@ -134,6 +103,7 @@ 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
@@ -141,11 +111,7 @@ with col2:
             if image_match:
                 st.session_state.image_url = image_match.group(0)
                 st.session_state.image_updated = True
-                st.rerun()
+                st.rerun()  # Trigger rerun to refresh image display
 
         except Exception as e:
             st.error(f"❌ Error: {str(e)}")
-
-    for message in reversed(st.session_state.messages):
-        role, content = message["role"], message["content"]
-        st.chat_message(role).write(content)