Update app.py

app.py

@@ -124,21 +124,20 @@ class SpeechAnalyzer:
         
         return aggregated_scores
 
-    def analyze_emotional_trajectory(self, text, window_size=5, ngram_size=3):
-        """Enhanced emotional trajectory analysis using efficient batch processing"""
-        segments = self.split_text(text, max_length=512)
+    def analyze_emotional_trajectory(self, text, window_size=5):
+        """Enhanced emotional trajectory analysis using sentence-level processing"""
+        segments = self.split_text(text, max_length=1024)
         sentiment_scores = []
         
-        # Process n-grams in larger batches while keeping full text
-        batch_size = 32
         for segment in segments:
-            words = segment.split()
-            ngrams = [' '.join(words[i:i+ngram_size]) for i in range(0, len(words)-ngram_size+1)]
+            # Split into sentences using spaCy or NLTK
+            sentences = nltk.sent_tokenize(segment)
             
-            # Process full ngrams in batches
+            # Process sentences in batches
+            batch_size = 64
             segment_scores = []
-            for i in range(0, len(ngrams), batch_size):
-                batch = ngrams[i:i+batch_size]
+            for i in range(0, len(sentences), batch_size):
+                batch = sentences[i:i+batch_size]
                 results = self.sentiment_pipeline(batch)
                 batch_scores = [result['score'] for result in results]
                 segment_scores.extend(batch_scores)
@@ -146,7 +145,7 @@ class SpeechAnalyzer:
             avg_score = np.mean(segment_scores) if segment_scores else 0
             sentiment_scores.append(avg_score)
         
-        # Normalize scores to preserve full range of sentiment
+        # Normalize scores
         if sentiment_scores:
             min_score = min(sentiment_scores)
             max_score = max(sentiment_scores)
@@ -156,6 +155,7 @@ class SpeechAnalyzer:
         
         return sentiment_scores
 
+
     def detect_named_entities(self, text):
         """Detect named entities in the text"""
         entities = self.ner_pipeline(text)
@@ -305,44 +305,21 @@ def main():
             # Detailed insights
             for foundation, score in moral_scores.items():
                 st.write(f"**{MORAL_FOUNDATIONS[foundation]}**: {score:.2%}")
-                
+        
         with tab2:
             status_text.text('Processing Emotional Trajectory...')
             progress_bar.progress(40)
             st.subheader("Speech Trajectory Analysis")
             col1, col2 = st.columns(2)
             
-            # First, create consistent segments for both analyses
-            segments = analyzer.split_text(text, max_length=512)
+            # Create consistent segments for both analyses
+            segments = analyzer.split_text(text, max_length=1024)
             num_segments = len(segments)
             segment_labels = [f"{i+1}" for i in range(num_segments)]
             
             with col1:
                 st.write("### Emotional Flow")
-                sentiment_scores = []
-                
-                for segment in segments:
-                    # Get words for n-gram analysis within each segment
-                    words = segment.split()
-                    ngram_size = 3
-                    ngrams = [' '.join(words[i:i+ngram_size]) for i in range(0, len(words)-ngram_size+1)]
-                    
-                    # Calculate segment score from n-grams
-                    segment_scores = []
-                    for ngram in ngrams:
-                        result = analyzer.sentiment_pipeline(ngram)[0]
-                        segment_scores.append(result['score'])
-                    
-                    # Use average score for the segment
-                    avg_score = np.mean(segment_scores) if segment_scores else 0
-                    sentiment_scores.append(avg_score)
-                
-                # Normalize scores to show relative variations
-                min_score = min(sentiment_scores)
-                max_score = max(sentiment_scores)
-                score_range = max_score - min_score
-                if score_range > 0:
-                    sentiment_scores = [(s - min_score) / score_range * 2 - 1 for s in sentiment_scores]
+                sentiment_scores = analyzer.analyze_emotional_trajectory(text)
                 
                 trajectory_fig = go.Figure(data=go.Scatter(
                     x=segment_labels,
@@ -402,7 +379,7 @@ def main():
                     )
                 )
                 st.plotly_chart(moral_fig)
-        
+
         with tab3:
             status_text.text('Analyzing Linguistic Features...')
             progress_bar.progress(60)