Royalflush Speaker Diarization System for ICASSP 2022 Multi-channel Multi-party Meeting Transcription Challenge
This work addresses speaker diarization for multi-party meetings, presenting incremental improvements through system fusion and specific techniques.
The paper tackles speaker diarization in multi-channel meeting recordings by proposing a system with overlapped speech detection, speech separation, and speaker embedding methods, achieving a reduction in diarization error rate from 15.25% to 6.30% on the Alimeeting evaluation set.
This paper describes the Royalflush speaker diarization system submitted to the Multi-channel Multi-party Meeting Transcription Challenge(M2MeT). Our system comprises speech enhancement, overlapped speech detection, speaker embedding extraction, speaker clustering, speech separation and system fusion. In this system, we made three contributions. First, we propose an architecture of combining the multi-channel and U-Net-based models, aiming at utilizing the benefits of these two individual architectures, for far-field overlapped speech detection. Second, in order to use overlapped speech detection model to help speaker diarization, a speech separation based overlapped speech handling approach, in which the speaker verification technique is further applied, is proposed. Third, we explore three speaker embedding methods, and obtained the state-of-the-art performance on the CNCeleb-E test set. With these proposals, our best individual system significantly reduces DER from 15.25% to 6.40%, and the fusion of four systems finally achieves a DER of 6.30% on the far-field Alimeeting evaluation set.