Simple and Effective Unsupervised Speech Synthesis
Sound
2022-04-21 v3 Computation and Language
Audio and Speech Processing
Abstract
We introduce the first unsupervised speech synthesis system based on a simple, yet effective recipe. The framework leverages recent work in unsupervised speech recognition as well as existing neural-based speech synthesis. Using only unlabeled speech audio and unlabeled text as well as a lexicon, our method enables speech synthesis without the need for a human-labeled corpus. Experiments demonstrate the unsupervised system can synthesize speech similar to a supervised counterpart in terms of naturalness and intelligibility measured by human evaluation.
Cite
@article{arxiv.2204.02524,
title = {Simple and Effective Unsupervised Speech Synthesis},
author = {Alexander H. Liu and Cheng-I Jeff Lai and Wei-Ning Hsu and Michael Auli and Alexei Baevski and James Glass},
journal= {arXiv preprint arXiv:2204.02524},
year = {2022}
}
Comments
preprint, equal contribution from first two authors