Update app.py

app.py

@@ -5,6 +5,7 @@ import torchaudio
 from audio_recorder_streamlit import audio_recorder
 import torch
 from io import BytesIO
+import hashlib
 
 # Load Whisper model (cached)
 @st.cache_resource
@@ -25,11 +26,13 @@ def process_audio(audio_bytes):
 st.title("Real-Time Voice Typing")
 st.write("Type or speak - text will appear instantly!")
 
-# Initialize text in session state
+# Initialize session state
 if 'text_input' not in st.session_state:
     st.session_state.text_input = ""
+if 'last_audio_hash' not in st.session_state:
+    st.session_state.last_audio_hash = ""
 
-# Main text area (auto-updates from session state)
+# Main text area
 text_input = st.text_area(
     "Your text will appear here:", 
     value=st.session_state.text_input,
@@ -39,32 +42,38 @@ text_input = st.text_area(
 
 # Audio recorder component
 audio_bytes = audio_recorder(
-    pause_threshold=2.0,  # Stop after 2 seconds of silence
+    pause_threshold=1.5,  # Shorter pause threshold
     text="Speak to type",
     recording_color="#e8b62c",
     neutral_color="#6aa36f",
 )
 
-# Process audio in real-time
+# Process new audio only if it's different from last time
 if audio_bytes:
-    try:
-        audio_input = process_audio(audio_bytes)
-        whisper = load_model()
-        transcribed_text = whisper(audio_input)["text"]
-        
-        # Append new transcription to existing text
-        st.session_state.text_input = st.session_state.text_input + " " + transcribed_text
-        st.rerun()  # Updated to use st.rerun() instead of experimental_rerun()
-        
-    except Exception as e:
-        st.error(f"Error: {str(e)}")
+    current_hash = hashlib.md5(audio_bytes).hexdigest()
+    if current_hash != st.session_state.last_audio_hash:
+        st.session_state.last_audio_hash = current_hash
+        try:
+            audio_input = process_audio(audio_bytes)
+            whisper = load_model()
+            transcribed_text = whisper(audio_input)["text"]
+            
+            # Append new transcription only if different
+            if (not st.session_state.text_input.endswith(transcribed_text.strip()) and 
+                len(transcribed_text.strip()) > 0):
+                st.session_state.text_input += " " + transcribed_text
+                st.rerun()
+                
+        except Exception as e:
+            st.error(f"Error: {str(e)}")
 
 # Control buttons
 col1, col2 = st.columns(2)
 with col1:
     if st.button("Clear Text"):
         st.session_state.text_input = ""
-        st.rerun()  # Updated here as well
+        st.session_state.last_audio_hash = ""
+        st.rerun()
 with col2:
     st.download_button(
         "Download Text",