Self-attention generative adversarial network for speech enhancement

Huy Phan*, Huy Le Nguyen, Oliver Y. Chén, Philipp Koch, Ngoc Q.K. Duong, Ian McLoughlin, Alfred Mertins

*Corresponding author for this work


Existing generative adversarial networks (GANs) for speech enhancement solely rely on the convolution operation, which may obscure temporal dependencies across the sequence input. To remedy this issue, we propose a self-attention layer adapted from non-local attention, coupled with the convolutional and deconvolutional layers of a speech enhancement GAN (SEGAN) using raw signal input. Further, we empirically study the effect of placing the self-attention layer at the (de)convolutional layers with varying layer indices as well as at all of them when memory allows. Our experiments show that introducing self-attention to SEGAN leads to consistent improvement across the objective evaluation metrics of enhancement performance. Furthermore, applying at different (de)convolutional layers does not significantly alter performance, suggesting that it can be conveniently applied at the highest-level (de)convolutional layer with the smallest memory overhead.

Original languageEnglish
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages (from-to)7103 - 7107
Publication statusPublished - 2021


Dive into the research topics of 'Self-attention generative adversarial network for speech enhancement'. Together they form a unique fingerprint.

Cite this