search

pyannote / pyannote-audio

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding
http://pyannote.github.io
MIT License
6.38k stars 784 forks source link

fix a bug than happens when less than min(self.batch_size, 9) examples are used in the validation #1792

Closed antoinelaurent closed 2 days ago

antoinelaurent commented 3 days ago

fix a bug than happens when less than min(self.batch_size, 9) examples are used in the validation, which is the case for the debug protocol

antoinelaurent commented 2 days ago

Should be OK now.