MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions

23 Feb 2023  ·  Shengkui Zhao, Bin Ma ·

Transformer based models have provided significant performance improvements in monaural speech separation. However, there is still a performance gap compared to a recent proposed upper bound. The major limitation of the current dual-path Transformer models is the inefficient modelling of long-range elemental interactions and local feature patterns. In this work, we achieve the upper bound by proposing a gated single-head transformer architecture with convolution-augmented joint self-attentions, named \textit{MossFormer} (\textit{Mo}naural \textit{s}peech \textit{s}eparation Trans\textit{Former}). To effectively solve the indirect elemental interactions across chunks in the dual-path architecture, MossFormer employs a joint local and global self-attention architecture that simultaneously performs a full-computation self-attention on local chunks and a linearised low-cost self-attention over the full sequence. The joint attention enables MossFormer model full-sequence elemental interaction directly. In addition, we employ a powerful attentive gating mechanism with simplified single-head self-attentions. Besides the attentive long-range modelling, we also augment MossFormer with convolutions for the position-wise local pattern modelling. As a consequence, MossFormer significantly outperforms the previous models and achieves the state-of-the-art results on WSJ0-2/3mix and WHAM!/WHAMR! benchmarks. Our model achieves the SI-SDRi upper bound of 21.2 dB on WSJ0-3mix and only 0.3 dB below the upper bound of 23.1 dB on WSJ0-2mix.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Speech Separation WHAM! MossFormer (L) + DM SI-SDRi 17.3 # 1
Speech Separation WHAMR! MossFormer (L) + DM SI-SDRi 16.3 # 1
Speech Separation WSJ0-2mix MossFormer (L) + DM SI-SDRi 22.8 # 2
Speech Separation WSJ0-2mix MossFormer (M) + DM SI-SDRi 22.5 # 3
Speech Separation WSJ0-3mix MossFormer (L) + DM SI-SDRi 21.2 # 1
Speech Separation WSJ0-3mix MossFormer (M) + DM SI-SDRi 20.8 # 3