Update voice_analysis.py

voice_analysis.py

@@ -23,14 +23,23 @@ def diarize_speakers(audio_path):
     pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1", use_auth_token=hf_token)
     diarization = pipeline(audio_path)
     
-    # Identify the most frequent speaker
+    # Identify the speakers and their segments
     speaker_segments = {}
     for turn, _, speaker in diarization.itertracks(yield_label=True):
         if speaker not in speaker_segments:
-            speaker_segments[speaker] = 0
-        speaker_segments[speaker] += turn.end - turn.start
+            speaker_segments[speaker] = []
+        speaker_segments[speaker].append((turn.start, turn.end))
+        
+        # Print each voice segment
+        print(f"Speaker {speaker}: {turn.start:.2f}s - {turn.end:.2f}s")
+    
+    print("\nSpeaker Summary:")
+    for speaker, segments in speaker_segments.items():
+        total_duration = sum(end - start for start, end in segments)
+        print(f"Speaker {speaker}: Total duration = {total_duration:.2f}s")
     
-    most_frequent_speaker = max(speaker_segments, key=speaker_segments.get)
+    most_frequent_speaker = max(speaker_segments, key=lambda k: sum(end-start for start, end in speaker_segments[k]))
+    print(f"\nMost frequent speaker: {most_frequent_speaker}")
     
     return diarization, most_frequent_speaker