Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech
Audio and Speech Processing
2021-05-04 v5 Machine Learning
Sound
Abstract
The few-shot multi-speaker multi-style voice cloning task is to synthesize utterances with voice and speaking style similar to a reference speaker given only a few reference samples. In this work, we investigate different speaker representations and proposed to integrate pretrained and learnable speaker representations. Among different types of embeddings, the embedding pretrained by voice conversion achieves the best performance. The FastSpeech 2 model combined with both pretrained and learnable speaker representations shows great generalization ability on few-shot speakers and achieved 2nd place in the one-shot track of the ICASSP 2021 M2VoC challenge.
Keywords
Cite
@article{arxiv.2103.04088,
title = {Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech},
author = {Chung-Ming Chien and Jheng-Hao Lin and Chien-yu Huang and Po-chun Hsu and Hung-yi Lee},
journal= {arXiv preprint arXiv:2103.04088},
year = {2021}
}
Comments
Accepted by ICASSP 2021, in the special session of ICASSP 2021 M2VoC Challenge