Update app.py

app.py

@@ -10,57 +10,13 @@ import io
 import matplotlib.pyplot as plt
 import librosa.display
 from PIL import Image  # For image conversion
-from datetime import datetime
 
-# ---------------------------
-# Firebase Setup
-# ---------------------------
-import firebase_admin
-from firebase_admin import credentials, db
-from google.cloud import storage
-
-# Update the path below to your Firebase service account key JSON file
-SERVICE_ACCOUNT_KEY = "serviceAccountKey.json"  # <-- Ensure this file exists in your project directory
-if not os.path.exists(SERVICE_ACCOUNT_KEY):
-    raise FileNotFoundError(f"Firebase credentials file {SERVICE_ACCOUNT_KEY} not found!")
-
-# Initialize Firebase Admin for Realtime Database
-cred = credentials.Certificate(SERVICE_ACCOUNT_KEY)
-firebase_admin.initialize_app(cred, {
-    'databaseURL': 'https://sentiment-analysis-7562e-default-rtdb.firebaseio.com/'  # <-- Update with your actual DB URL
-})
-
-def upload_file_to_firebase(file_path, destination_blob_name):
-    """
-    Uploads a file to Firebase Storage and returns its public URL.
-    """
-    # Update the bucket name to match your Firebase Storage bucket (usually: your-project-id.appspot.com)
-    bucket_name = "sentiment-analysis-7562e.appspot.com"  # <-- Update with your storage bucket name
-    storage_client = storage.Client.from_service_account_json(SERVICE_ACCOUNT_KEY)
-    bucket = storage_client.bucket(bucket_name)
-    blob = bucket.blob(destination_blob_name)
-    blob.upload_from_filename(file_path)
-    blob.make_public()
-    print(f"File uploaded to {blob.public_url}")
-    return blob.public_url
-
-def store_prediction_metadata(file_url, predicted_emotion):
-    """
-    Stores the file URL, predicted emotion, and timestamp in Firebase Realtime Database.
-    """
-    ref = db.reference('predictions')
-    data = {
-        'file_url': file_url,
-        'predicted_emotion': predicted_emotion,
-        'timestamp': datetime.now().isoformat()
-    }
-    new_record_ref = ref.push(data)
-    print(f"Stored metadata with key: {new_record_ref.key}")
-    return new_record_ref.key
-
-# ---------------------------
-# Emotion Recognition Code
-# ---------------------------
+# Try to import noisereduce (if not available, noise reduction will be skipped)
+try:
+    import noisereduce as nr
+    NOISEREDUCE_AVAILABLE = True
+except ImportError:
+    NOISEREDUCE_AVAILABLE = False
 
 # Mapping from emotion labels to emojis
 emotion_to_emoji = {
@@ -79,7 +35,7 @@ def add_emoji_to_label(label):
     emoji = emotion_to_emoji.get(label.lower(), "")
     return f"{label.capitalize()} {emoji}"
 
-# Load the pre-trained SpeechBrain classifier (Emotion Recognition with wav2vec2 on IEMOCAP)
+# Load the pre-trained SpeechBrain classifier
 classifier = foreign_class(
     source="speechbrain/emotion-recognition-wav2vec2-IEMOCAP",
     pymodule_file="custom_interface.py",
@@ -96,11 +52,6 @@ def preprocess_audio(audio_file, apply_noise_reduction=False):
     Saves the processed audio to a temporary file and returns its path.
     """
     y, sr = librosa.load(audio_file, sr=16000, mono=True)
-    try:
-        import noisereduce as nr
-        NOISEREDUCE_AVAILABLE = True
-    except ImportError:
-        NOISEREDUCE_AVAILABLE = False
     if apply_noise_reduction and NOISEREDUCE_AVAILABLE:
         y = nr.reduce_noise(y=y, sr=sr)
     if np.max(np.abs(y)) > 0:
@@ -159,7 +110,7 @@ def predict_emotion(audio_file, use_ensemble=False, apply_noise_reduction=False,
             result = classifier.classify_file(temp_file)
             os.remove(temp_file)
             if isinstance(result, tuple) and len(result) > 3:
-                label = result[3][0]
+                label = result[3][0]  # Extract predicted emotion label from the tuple
             else:
                 label = str(result)
         return add_emoji_to_label(label.lower())
@@ -183,52 +134,39 @@ def plot_waveform(audio_file):
 def predict_and_plot(audio_file, use_ensemble, apply_noise_reduction, segment_duration, overlap):
     """
     Run emotion prediction and generate a waveform plot.
-    Additionally, upload the audio file to Firebase Storage and store the metadata in Firebase Realtime Database.
     Returns a tuple: (emotion label with emoji, waveform image as a PIL Image).
     """
-    # Upload the original audio file to Firebase Storage
-    destination_blob_name = os.path.basename(audio_file)
-    file_url = upload_file_to_firebase(audio_file, destination_blob_name)
-    
-    # Predict emotion and generate waveform
     emotion = predict_emotion(audio_file, use_ensemble, apply_noise_reduction, segment_duration, overlap)
     waveform = plot_waveform(audio_file)
-    
-    # Store metadata (file URL and predicted emotion) in Firebase Realtime Database
-    record_key = store_prediction_metadata(file_url, emotion)
-    print(f"Record stored with key: {record_key}")
-    
     return emotion, waveform
 
-# ---------------------------
-# Gradio App UI
-# ---------------------------
 with gr.Blocks(css=".gradio-container {background-color: #f7f7f7; font-family: Arial;}") as demo:
     gr.Markdown("<h1 style='text-align: center;'>Enhanced Emotion Recognition</h1>")
     gr.Markdown(
         "Upload an audio file, and the model will predict the emotion using a wav2vec2 model fine-tuned on IEMOCAP data. "
-        "The prediction is accompanied by an emoji, and you can view the audio's waveform. "
-        "The audio file and prediction metadata are stored in Firebase Realtime Database."
+        "The prediction is accompanied by an emoji in the output, and you can also view the audio's waveform. "
+        "Use the options below to adjust ensemble prediction and noise reduction settings."
     )
     
     with gr.Tabs():
         with gr.TabItem("Emotion Recognition"):
             with gr.Row():
                 audio_input = gr.Audio(type="filepath", label="Upload Audio")
-            use_ensemble_checkbox = gr.Checkbox(label="Use Ensemble Prediction (for long audio)", value=False)
-            apply_noise_reduction_checkbox = gr.Checkbox(label="Apply Noise Reduction", value=False)
+            use_ensemble = gr.Checkbox(label="Use Ensemble Prediction (for long audio)", value=False)
+            apply_noise_reduction = gr.Checkbox(label="Apply Noise Reduction", value=False)
             with gr.Row():
-                segment_duration_slider = gr.Slider(minimum=1.0, maximum=10.0, step=0.5, value=3.0, label="Segment Duration (s)")
-                overlap_slider = gr.Slider(minimum=0.0, maximum=5.0, step=0.5, value=1.0, label="Segment Overlap (s)")
+                segment_duration = gr.Slider(minimum=1.0, maximum=10.0, step=0.5, value=3.0, label="Segment Duration (s)")
+                overlap = gr.Slider(minimum=0.0, maximum=5.0, step=0.5, value=1.0, label="Segment Overlap (s)")
             predict_button = gr.Button("Predict Emotion")
             result_text = gr.Textbox(label="Predicted Emotion")
             waveform_image = gr.Image(label="Audio Waveform", type="pil")
             
             predict_button.click(
                 predict_and_plot,
-                inputs=[audio_input, use_ensemble_checkbox, apply_noise_reduction_checkbox, segment_duration_slider, overlap_slider],
+                inputs=[audio_input, use_ensemble, apply_noise_reduction, segment_duration, overlap],
                 outputs=[result_text, waveform_image]
             )
+        
         with gr.TabItem("About"):
             gr.Markdown("""
 **Enhanced Emotion Recognition App**
@@ -238,8 +176,7 @@ with gr.Blocks(css=".gradio-container {background-color: #f7f7f7; font-family: A
   - Ensemble Prediction for long audio files.
   - Optional Noise Reduction.
   - Visualization of the audio waveform.
-  - Emoji representation of the predicted emotion.
-  - Audio file and prediction metadata stored in Firebase Realtime Database.
+  - Emoji representation of the predicted emotion in the output.
 
 **Credits:**
 - [SpeechBrain](https://speechbrain.github.io)
@@ -247,4 +184,4 @@ with gr.Blocks(css=".gradio-container {background-color: #f7f7f7; font-family: A
             """)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()