Learning Bandwidth Expansion Using Perceptually-motivated Loss
Lists
Only you see this
Abstract
We introduce a perceptually motivated approach to bandwidth expansion for speech. Our method pairs a new 3-way split variant of the FFTNet neural vocoder structure with a perceptual loss function, combining objectives from both the time and frequency domains. Mean opinion score tests show that it outperforms baseline methods from both domains, even for extreme bandwidth expansion.
BibTeX
@inproceedings{feng2019learning,
author = {Berthy Feng and Zeyu Jin and Jiaqi Su and Adam Finkelstein},
title = {{Learning Bandwidth Expansion Using Perceptually-motivated Loss}},
booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
year = {2019},
doi = {10.1109/icassp.2019.8682367},
}Loading related content...