Update app.py

app.py

@@ -6,11 +6,24 @@ import speech_recognition as sr
 import tempfile
 from audio_recorder_streamlit import audio_recorder
 import numpy as np
+import wave
+import io
 
 # Load sentiment analysis model using PyTorch backend
 mood_classifier = pipeline("sentiment-analysis", framework="pt")
 
-# Functions
+def convert_audio_to_wav(audio_bytes):
+    # Create a wave file in memory
+    wav_buffer = io.BytesIO()
+    
+    with wave.open(wav_buffer, 'wb') as wav_file:
+        wav_file.setnchannels(1)  # Mono
+        wav_file.setsampwidth(2)  # 2 bytes per sample
+        wav_file.setframerate(44100)  # Sample rate
+        wav_file.writeframes(audio_bytes)
+    
+    return wav_buffer.getvalue()
+
 def detect_mood(text):
     result = mood_classifier(text)[0]
     if result['label'] == 'POSITIVE':
@@ -24,36 +37,42 @@ def speech_to_text():
     # Initialize recognizer
     r = sr.Recognizer()
     
-    # Create a temporary file to store the recorded audio
-    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as fp:
-        # Add audio recording widget
-        audio_bytes = audio_recorder(
-            text="Click to record",
-            recording_color="#e8b62c",
-            neutral_color="#6aa36f"
-        )
-        
-        if audio_bytes:
-            # Save audio bytes to temporary file
-            fp.write(audio_bytes)
-            temp_filename = fp.name
-            
-            # Read the audio file
-            with sr.AudioFile(temp_filename) as source:
-                # Adjust for ambient noise and record
-                r.adjust_for_ambient_noise(source)
-                audio = r.record(source)
+    # Add audio recording widget
+    audio_bytes = audio_recorder(
+        text="Click to record your mood",
+        recording_color="#e8b62c",
+        neutral_color="#6aa36f",
+        pause_threshold=2.0  # Automatically stop after 2 seconds of silence
+    )
+    
+    if audio_bytes:
+        try:
+            # Create a temporary WAV file
+            with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as temp_audio:
+                # Convert and write audio bytes to WAV format
+                wav_bytes = convert_audio_to_wav(audio_bytes)
+                temp_audio.write(wav_bytes)
+                temp_audio.flush()
                 
-                try:
-                    # Use Google Speech Recognition
-                    text = r.recognize_google(audio)
-                    return text
-                except sr.UnknownValueError:
-                    st.error("Could not understand the audio")
-                    return None
-                except sr.RequestError:
-                    st.error("Could not request results from speech recognition service")
-                    return None
+                # Use the temporary file for speech recognition
+                with sr.AudioFile(temp_audio.name) as source:
+                    # Record the audio file
+                    audio = r.record(source)
+                    
+                    try:
+                        # Attempt speech recognition
+                        text = r.recognize_google(audio)
+                        st.success("Speech recognized successfully!")
+                        return text
+                    except sr.UnknownValueError:
+                        st.error("Could not understand the audio. Please try speaking clearly and try again.")
+                        return None
+                    except sr.RequestError as e:
+                        st.error(f"Could not request results from speech recognition service; {e}")
+                        return None
+        except Exception as e:
+            st.error(f"Error processing audio: {e}")
+            return None
     return None
 
 def get_song_recommendations(mood, api_key):
@@ -111,12 +130,13 @@ if input_method == "Text":
     user_text = user_mood
 else:
     # Speech input
-    st.write("📢 Tell me about your day...")
+    st.write("📢 Click the button below and tell me about your day...")
     user_text = speech_to_text()
     
     if user_text:
-        st.write(f"You said: {user_text}")
+        st.info(f"You said: '{user_text}'")
         user_mood = detect_mood(user_text)
+        st.write(f"Based on what you said, I detect your mood as: **{user_mood}**")
 
 # Playlist
 if 'playlist' not in st.session_state: