Continuous speech separation plays a vital role in complicated speech related tasks such as conversation transcription. The separation model extracts a single speaker signal from a mixed speech. In this paper, we use transformer and conformer in lieu of recurrent neural networks in the separation system, as we believe capturing global information with the self-attention based method is crucial for the speech separation. Evaluating on the LibriCSS dataset, the conformer separation model achieves state of the art results, with a relative 23.5% word error rate (WER) reduction from bi-directional LSTM (BLSTM) in the utterance-wise evaluation and a 15.4% WER reduction in the continuous evaluation.
@article{arxiv.2008.05773,
title = {Continuous Speech Separation with Conformer},
author = {Sanyuan Chen and Yu Wu and Zhuo Chen and Jian Wu and Jinyu Li and Takuya Yoshioka and Chengyi Wang and Shujie Liu and Ming Zhou},
journal= {arXiv preprint arXiv:2008.05773},
year = {2020}
}