Chronological Self-Training for Real-Time Speaker Diarization
This addresses the need for efficient user interaction in real-time diarization systems, though it appears incremental as it builds on existing self-training methods.
The paper tackled the problem of real-time speaker diarization with limited enrollment samples by proposing a chronological self-training approach, achieving over 95% accuracy with just 1 second of training and average diarization error rates as low as 10% on multilingual audio data.
Diarization partitions an audio stream into segments based on the voices of the speakers. Real-time diarization systems that include an enrollment step should limit enrollment training samples to reduce user interaction time. Although training on a small number of samples yields poor performance, we show that the accuracy can be improved dramatically using a chronological self-training approach. We studied the tradeoff between training time and classification performance and found that 1 second is sufficient to reach over 95% accuracy. We evaluated on 700 audio conversation files of about 10 minutes each from 6 different languages and demonstrated average diarization error rates as low as 10%.