Update app.py

app.py

@@ -1,75 +1,132 @@
 import gradio as gr
-from transformers import pipeline
-import torch
+import numpy as np
+from textblob import TextBlob
+import speech_recognition as sr
 
-def create_analyzers():
-    """Initialize speech and emotion analyzers"""
-    try:
-        # Use tiny whisper model for speed and reliability
-        speech_recognizer = pipeline(
-            "automatic-speech-recognition",
-            model="openai/whisper-tiny.en",
-            chunk_length_s=30
-        )
-        
-        # Use smaller emotion classifier
-        emotion_classifier = pipeline(
-            "text-classification",
-            model="SamLowe/roberta-base-go_emotions",
-            top_k=1
-        )
-        
-        return speech_recognizer, emotion_classifier
-    except Exception as e:
-        print(f"Model loading error: {e}")
-        return None, None
+class SentimentAnalyzer:
+    def __init__(self):
+        self.recognizer = sr.Recognizer()
+    
+    def audio_to_text(self, audio):
+        """Convert audio to text using speech recognition"""
+        try:
+            # Get audio data from Gradio input
+            sample_rate, audio_data = audio
+            
+            # Convert audio data to audio file format that speech_recognition can use
+            import io
+            import scipy.io.wavfile as wav
+            byte_io = io.BytesIO()
+            wav.write(byte_io, sample_rate, audio_data.astype(np.int16))
+            byte_io.seek(0)
+            
+            # Use speech recognition
+            with sr.AudioFile(byte_io) as source:
+                audio_data = self.recognizer.record(source)
+                text = self.recognizer.recognize_google(audio_data)
+                return text
+        except Exception as e:
+            return f"Error in speech recognition: {str(e)}"
+    
+    def analyze_sentiment(self, text):
+        """Analyze sentiment using TextBlob"""
+        try:
+            blob = TextBlob(text)
+            # Get polarity (-1 to 1) and subjectivity (0 to 1)
+            polarity = blob.sentiment.polarity
+            subjectivity = blob.sentiment.subjectivity
+            
+            # Determine sentiment category
+            if polarity > 0:
+                sentiment = "Positive"
+            elif polarity < 0:
+                sentiment = "Negative"
+            else:
+                sentiment = "Neutral"
+            
+            # Format results
+            results_text = f"""
+Detected Text: "{text}"
 
-def analyze_tone(audio_file):
-    """Analyze the emotional tone of speech"""
-    if audio_file is None:
-        return "No input", "N/A"
+Analysis Results:
+- Overall Sentiment: {sentiment}
+- Polarity Score: {polarity:.2f} (-1 to +1)
+- Subjectivity Score: {subjectivity:.2f} (0 to 1)
+"""
+            
+            # Prepare plot data
+            plot_data = {
+                "labels": ["Polarity", "Subjectivity"],
+                "values": [polarity * 100, subjectivity * 100]  # Convert to percentage for visualization
+            }
+            
+            return results_text, plot_data
+            
+        except Exception as e:
+            return f"Error in sentiment analysis: {str(e)}", None
+
+def create_interface():
+    analyzer = SentimentAnalyzer()
     
-    try:
-        # Get models
-        speech_recognizer, emotion_classifier = create_analyzers()
+    def process_audio(audio):
+        if audio is None:
+            return "Please provide an audio input.", None
         
-        # Transcribe audio
-        transcription = speech_recognizer(audio_file)
-        text = transcription["text"]
-        
-        if not text.strip():
-            return "No speech detected", "N/A"
-            
-        print(f"Transcribed text: {text}")  # For debugging
+        # Convert audio to text
+        text = analyzer.audio_to_text(audio)
+        if text.startswith("Error"):
+            return text, None
         
-        # Analyze emotion
-        result = emotion_classifier(text)[0][0]
-        emotion = result['label'].replace('_', ' ').title()
-        confidence = f"{result['score']:.1%}"
+        # Analyze sentiment
+        return analyzer.analyze_sentiment(text)
+
+    # Create Gradio interface
+    with gr.Blocks() as interface:
+        gr.Markdown("# 🎤 Speech Sentiment Analysis")
+        gr.Markdown("""
+        Speak or upload an audio file to analyze its emotional content.
+        The system will convert speech to text and analyze the sentiment.
+        """)
         
-        return emotion, confidence
+        with gr.Row():
+            with gr.Column():
+                audio_input = gr.Audio(
+                    label="Upload or Record Audio",
+                    type="numpy",
+                    sources=["microphone", "upload"]
+                )
+                analyze_btn = gr.Button("Analyze Sentiment")
+            
+            with gr.Column():
+                output_text = gr.Textbox(
+                    label="Analysis Results",
+                    lines=8
+                )
+                output_plot = gr.BarPlot(
+                    title="Sentiment Scores",
+                    x_title="Metrics",
+                    y_title="Score (%)"
+                )
         
-    except Exception as e:
-        print(f"Analysis error: {e}")
-        return f"Error: {str(e)}", "N/A"
-
-# Create minimal interface
-interface = gr.Interface(
-    fn=analyze_tone,
-    inputs=[
-        gr.Audio(
-            sources=["microphone", "upload"],
-            type="filepath",
-            label="Audio Input"
+        analyze_btn.click(
+            fn=process_audio,
+            inputs=[audio_input],
+            outputs=[output_text, output_plot]
         )
-    ],
-    outputs=[
-        gr.Textbox(label="Emotion"),
-        gr.Textbox(label="Confidence")
-    ],
-    title="Speech Emotion Analyzer",
-    description="Record or upload audio to detect the emotional tone.",
-)
+        
+        gr.Markdown("""
+        ### How to Use:
+        1. Click the microphone button to record or upload an audio file
+        2. Click "Analyze Sentiment" to process
+        3. View the results showing:
+           - Detected text from speech
+           - Overall sentiment (Positive/Negative/Neutral)
+           - Polarity score (-100% to +100%)
+           - Subjectivity score (0% to 100%)
+        """)
+    
+    return interface
 
 if __name__ == "__main__":
-    interface.launch(server_name="0.0.0.0", share=True)
+    demo = create_interface()
+    demo.launch(share=True)