Local spectral attention for full-band speech enhancement.

Abstract

Attention mechanism has been widely used in speech enhancement (SE) because, theoretically, it can effectively model the inherent connection of signal in time domain and spectrum domain. In this Letter, it is found that the attention over the entire frequency range hampers the inference for full-band SE and possibly leads to excessive residual noise and degradation of speech. To alleviate this problem, the local spectral attention is introduced into full-band SE model by limiting the span of attention. The ablation tests on three full-band SE models reveal that the local frequency attention can effectively improve overall performance.