Efficient Trainable Front-Ends for Neural Speech Enhancement
Audio and Speech Processing
2020-02-24 v1 Machine Learning
Neural and Evolutionary Computing
Sound
Machine Learning
Abstract
Many neural speech enhancement and source separation systems operate in the time-frequency domain. Such models often benefit from making their Short-Time Fourier Transform (STFT) front-ends trainable. In current literature, these are implemented as large Discrete Fourier Transform matrices; which are prohibitively inefficient for low-compute systems. We present an efficient, trainable front-end based on the butterfly mechanism to compute the Fast Fourier Transform, and show its accuracy and efficiency benefits for low-compute neural speech enhancement models. We also explore the effects of making the STFT window trainable.
Cite
@article{arxiv.2002.09286,
title = {Efficient Trainable Front-Ends for Neural Speech Enhancement},
author = {Jonah Casebeer and Umut Isik and Shrikant Venkataramani and Arvindh Krishnaswamy},
journal= {arXiv preprint arXiv:2002.09286},
year = {2020}
}
Comments
5 pages, 5 figures, ICASSP 2020