Update voice_input.py

voice_input.py

@@ -1,105 +1,65 @@
-import streamlit as st
-from typing import Dict, Optional
-
-def render_voice_examples(language_option: str) -> None:
-    """Render voice input examples based on language"""
-    if language_option == "Arabic":
-        st.markdown("""
-        جرب أن تقول شيئاً مثل:
-        > "أحتاج إلى طلب 500 ريال للمشروع 223 المسمى جامعة أبها لشراء بعض الأدوات"
-        """)
-    else:
-        st.markdown("""
-        Try saying something like:
-        > "I need to request 500 riyals for project 223 named Abha University to buy some tools"
-        """)
-
-def handle_voice_input(voice_handler, language: str, memory_handler, gemini_processor) -> Optional[Dict]:
-    """Handle voice input and processing"""
-    with st.spinner("Checking microphone access..."):
-        if not voice_handler.check_microphone_access():
-            st.error("""
-            Could not access microphone. Please ensure:
-            1. You're using a secure (HTTPS) connection
-            2. You've granted microphone permissions in your browser
-            3. Your microphone is properly connected and working
-            
-            Try refreshing the page and allowing microphone access when prompted.
-            """)
-            return None
-    
-    with st.spinner("Listening... Please speak clearly"):
-        voice_text = voice_handler.listen_for_voice(language)
-        
-        if voice_text.startswith("Error:"):
-            st.error(voice_text)
-            return None
-        
-        if voice_text.startswith("Could not"):
-            st.warning(voice_text)
-            return None
-        
-        st.success("Voice captured!")
-        st.write("You said:", voice_text)
-        
-        # Add to memory
-        memory_handler.add_interaction(voice_text)
-        
-        with st.spinner("Processing voice input..."):
-            context = memory_handler.get_context()
-            details = gemini_processor.extract_request_details(voice_text, context)
-            
-            if not details:
-                st.error("Could not extract request details. Please try again or use manual input.")
-                return None
-            
-            if 'translated_text' in details:
-                st.info(f"Translated text: {details['translated_text']}")
-            
-            if details.get('missing_fields'):
-                missing = ", ".join(details['missing_fields'])
-                st.warning(f"Please provide the following missing information: {missing}")
-            else:
-                st.success("Voice input processed! Please verify the details below.")
-            
-            return details
-
-def render_voice_input(voice_handler, language_option: str, language_mapping: Dict[str, str], 
-                      memory_handler, gemini_processor) -> None:
-    """Render the voice input section"""
-    if not voice_handler.permission_granted:
-        st.info("Microphone access is required for voice input. Click below to enable it.")
-        if st.button("Enable Microphone Access"):
-            if voice_handler.request_permissions():
-                st.success("Microphone access granted! You can now use voice input.")
-                st.rerun()
-            else:
-                st.error("""
-                Could not access microphone. Please:
-                1. Check if your browser blocks microphone access
-                2. Allow microphone access in your browser settings
-                3. Ensure your microphone is properly connected
-                """)
-    
-    col1, col2, col3 = st.columns([3, 1, 1])
-    
-    with col1:
-        render_voice_examples(language_option)
-    
-    with col2:
-        if st.button("🎤 Start Voice Input"):
-            details = handle_voice_input(
-                voice_handler, 
-                language_mapping[language_option],
-                memory_handler,
-                gemini_processor
-            )
-            if details:
-                st.session_state['voice_details'] = details
-    
-    with col3:
-        if st.button("🗑️ Clear Memory"):
-            memory_handler.clear_memory()
-            if 'voice_details' in st.session_state:
-                del st.session_state['voice_details']
-            st.success("Memory cleared!")
+import streamlit as st
+from voice_handler import VoiceHandler
+from memory_handler import MemoryHandler
+from gemini_processor import GeminiProcessor
+
+def render_voice_input(voice_handler, language_option, language_mapping, memory_handler, gemini_processor):
+    """Render the voice input section with improved error handling"""
+    
+    # Display voice input instructions based on language
+    if language_option == "Arabic":
+        st.markdown("""
+        ### تعليمات الإدخال الصوتي:
+        1. اضغط على زر "بدء التسجيل"
+        2. تحدث بوضوح في الميكروفون
+        3. سيتم معالجة كلامك تلقائياً
+        """)
+    else:
+        st.markdown("""
+        ### Voice Input Instructions:
+        1. Click the "Start Recording" button
+        2. Speak clearly into your microphone
+        3. Your speech will be processed automatically
+        """)
+    
+    # Check microphone access before showing the record button
+    mic_status, mic_message = voice_handler.check_microphone_access()
+    if not mic_status:
+        st.error(mic_message)
+        st.markdown("""
+        ### 🎤 Microphone Access Required
+        
+        Please ensure:
+        1. Your browser allows microphone access
+        2. Your microphone is properly connected
+        3. No other application is using the microphone
+        
+        To grant microphone access:
+        1. Click the lock/info icon in your browser's address bar
+        2. Allow microphone access for this site
+        3. Refresh the page after granting access
+        """)
+        return
+    
+    col1, col2 = st.columns([1, 4])
+    with col1:
+        if st.button("🎤 Start Recording"):
+            selected_language = language_mapping.get(language_option, "en-US")
+            with st.spinner("Listening..."):
+                voice_input = voice_handler.listen_for_voice(selected_language)
+                
+                if not voice_input.startswith("Error") and not voice_input.startswith("Could not"):
+                    memory_handler.add_interaction(voice_input)
+                    with st.spinner("Processing voice input..."):
+                        context = memory_handler.get_context()
+                        details = gemini_processor.extract_request_details(voice_input, context)
+                        
+                        if details:
+                            st.session_state['voice_details'] = details
+                            if 'translated_text' in details:
+                                st.info(f"Translated text: {details['translated_text']}")
+                            if details.get('missing_fields'):
+                                missing = ", ".join(details['missing_fields'])
+                                st.warning(f"Please provide the following missing information: {missing}")
+                            else:
+                                st.success("Voice input processed! Please verify the details below.")