English

Speech Denoising in the Waveform Domain with Self-Attention

Sound 2022-07-08 v3 Machine Learning Audio and Speech Processing

Abstract

In this work, we present CleanUNet, a causal speech denoising model on the raw waveform. The proposed model is based on an encoder-decoder architecture combined with several self-attention blocks to refine its bottleneck representations, which is crucial to obtain good results. The model is optimized through a set of losses defined over both waveform and multi-resolution spectrograms. The proposed method outperforms the state-of-the-art models in terms of denoised speech quality from various objective and subjective evaluation metrics. We release our code and models at https://github.com/nvidia/cleanunet.

Keywords

Cite

@article{arxiv.2202.07790,
  title  = {Speech Denoising in the Waveform Domain with Self-Attention},
  author = {Zhifeng Kong and Wei Ping and Ambrish Dantrey and Bryan Catanzaro},
  journal= {arXiv preprint arXiv:2202.07790},
  year   = {2022}
}

Comments

Published in ICASSP 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Listen to audio samples from CleanUNet at: https://cleanunet.github.io/

R2 v1 2026-06-24T09:40:03.113Z