Papers › Continuous Speech Separation with Conformer

Continuous Speech Separation with Conformer

13 Aug 2020arXiv:2008.05773archive 2025-07-28

Sanyuan Chen, Yu Wu, Zhuo Chen, Jian Wu, Jinyu Li, Takuya Yoshioka, Chengyi Wang, Shujie Liu, Ming Zhou

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.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

Sanyuan-Chen/CSS_with_Conformer mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Speech Separation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Separation LibriCSS Conformer (large) 0L 5.0 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (large) 0S 5.4 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (large) 10% 7.5 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (large) 20% 10.7 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (large) 30% 13.8 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (large) 40% 17.1 #1 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 0L 5.4 #2 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 0S 5.6 #2 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 10% 8.2 #2 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 20% 11.8 #2 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 30% 15.5 #2 of 2 Archive leaderboard report
Speech Separation LibriCSS Conformer (base) 40% 18.9 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

LSTMSigmoid ActivationTanh Activation

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections