Update state.py

state.py

@@ -719,16 +719,32 @@ def analyze(inFileName):
             noVoice, oneVoice, multiVoice = su.calcSpeakingTypes(pipeline, currAnnotation, currTotalTime)
         except Exception as e:
             print(f"calcSpeakingTypes failed ({e}), falling back to annotation-based voice split")
-            # Fall back: build a clean annotation containing only properly named
-            # speakers (non-None, non-empty) so build_df5 never receives None labels.
-            from pyannote.core import Annotation
+            # Fall back: derive voice categories directly from segment timeline.
+            # - oneVoice: each named speaker segment
+            # - noVoice:  gaps between segments (real silence)
+            # - multiVoice: empty (can't detect overlap without the classifier)
+            from pyannote.core import Annotation, Segment
             noVoice    = Annotation()
             multiVoice = Annotation()
             oneVoice   = Annotation()
-            for label in currAnnotation.labels():
-                if label is not None and str(label).strip() != "":
-                    for seg in currAnnotation.subset([label]).itersegments():
-                        oneVoice[seg] = label
+            # Sort all segments by start time
+            all_segs = sorted(
+                [(seg.start, seg.end, label)
+                 for label in currAnnotation.labels()
+                 if label is not None and str(label).strip() != ""
+                 for seg in currAnnotation.subset([label]).itersegments()],
+                key=lambda x: x[0]
+            )
+            # Fill oneVoice and detect gaps for noVoice
+            prev_end = 0.0
+            for start, end, label in all_segs:
+                if start > prev_end + 0.1:
+                    noVoice[Segment(prev_end, start)] = 'silence'
+                oneVoice[Segment(start, end)] = label
+                prev_end = max(prev_end, end)
+            # Trailing silence to end of file
+            if currTotalTime > prev_end + 0.1:
+                noVoice[Segment(prev_end, currTotalTime)] = 'silence' 
         sumNoVoice    = su.sumTimes(noVoice)
         sumOneVoice   = su.sumTimes(oneVoice)
         sumMultiVoice = su.sumTimes(multiVoice)