Update asr_diarization/pipeline.py

asr_diarization/pipeline.py

@@ -185,13 +185,6 @@ class ASR_Diarization:
             for t, _, spk in diarization.itertracks(yield_label=True)
         ]
         
-        print(f"DEBUG DIARIZATION:")
-        print(f"   Raw diarization segments: {len(diar_segments)}")
-        
-        # Count unique speakers BEFORE any processing
-        raw_speakers = list(set([seg['speaker'] for seg in diar_segments]))
-        print(f"   Raw unique speakers: {len(raw_speakers)} - {raw_speakers}")
-        
         # Step 2: Calculate SNR for adaptive processing
         snr = self.calculate_snr(audio_path)
         
@@ -227,24 +220,6 @@ class ASR_Diarization:
         print(f"Final: {len(filtered_segments)} segments for Whisper")
         return filtered_segments
 
-    def map_speaker_labels(self, segments, original_speakers=['A', 'B', 'C', 'D']):
-        """Map SPEAKER_XX labels to A, B, C, D format to match original"""
-        unique_speakers = list(set([seg['speaker'] for seg in segments]))
-        speaker_map = {}
-        
-        # Create mapping from SPEAKER_00 -> A, SPEAKER_01 -> B, etc.
-        for i, spk in enumerate(sorted(unique_speakers)):
-            if i < len(original_speakers):
-                speaker_map[spk] = original_speakers[i]
-            else:
-                speaker_map[spk] = f"SPK_{i}"
-        
-        # Apply mapping to all segments
-        for seg in segments:
-            seg['speaker'] = speaker_map[seg['speaker']]
-        
-        return segments, list(speaker_map.values())
-
     def merge_consecutive_speaker_segments(self, segments):
         """Merge only consecutive segments from the same speaker while preserving order"""
         if not segments:
@@ -384,9 +359,6 @@ class ASR_Diarization:
         # Merge consecutive segments by same speaker
         merged_segments = self.merge_consecutive_speaker_segments(merged_segments)
 
-        # Map speaker labels to match original format (A, B, C, D)
-        merged_segments, speakers = self.map_speaker_labels(merged_segments)
-
         # Combine ASR segments with NSE events if provided
         if nse_events:
             print(f"Combining {len(merged_segments)} ASR segments with {len(nse_events)} NSE events")