English

SCDNet: Self-supervised Learning Feature-based Speaker Change Detection

Audio and Speech Processing 2024-06-13 v1 Sound

Abstract

Speaker Change Detection (SCD) is to identify boundaries among speakers in a conversation. Motivated by the success of fine-tuning wav2vec 2.0 models for the SCD task, a further investigation of self-supervised learning (SSL) features for SCD is conducted in this work. Specifically, an SCD model, named SCDNet, is proposed. With this model, various state-of-the-art SSL models, including Hubert, wav2vec 2.0, and WavLm are investigated. To discern the most potent layer of SSL models for SCD, a learnable weighting method is employed to analyze the effectiveness of intermediate representations. Additionally, a fine-tuning-based approach is also implemented to further compare the characteristics of SSL models in the SCD task. Furthermore, a contrastive learning method is proposed to mitigate the overfitting tendencies in the training of both the fine-tuning-based method and SCDNet. Experiments showcase the superiority of WavLm in the SCD task and also demonstrate the good design of SCDNet.

Keywords

Cite

@article{arxiv.2406.08393,
  title  = {SCDNet: Self-supervised Learning Feature-based Speaker Change Detection},
  author = {Yue Li and Xinsheng Wang and Li Zhang and Lei Xie},
  journal= {arXiv preprint arXiv:2406.08393},
  year   = {2024}
}
R2 v1 2026-06-28T17:03:23.985Z