SVSNet: An End-to-end Speaker Voice Similarity Assessment Model

Neural evaluation metrics derived for numerous speech generation tasks have recently attracted great attention. In this paper, we propose SVSNet, the first end-to-end neural network model to assess the speaker voice similarity between converted speech and natural speech for voice conversion tasks. U...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Cheng-Hung, Hu, Yu-Huai Peng, Yamagishi, Junichi, Tsao, Yu, Wang, Hsin-Min
Format Paper Journal Article
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 17.02.2022
Subjects
Coders
Computer Science - Learning
Computer Science - Sound
Evaluation
Neural networks
Similarity
Speech recognition
Waveforms
Online AccessGet full text

Cover

More Information
Summary:Neural evaluation metrics derived for numerous speech generation tasks have recently attracted great attention. In this paper, we propose SVSNet, the first end-to-end neural network model to assess the speaker voice similarity between converted speech and natural speech for voice conversion tasks. Unlike most neural evaluation metrics that use hand-crafted features, SVSNet directly takes the raw waveform as input to more completely utilize speech information for prediction. SVSNet consists of encoder, co-attention, distance calculation, and prediction modules and is trained in an end-to-end manner. The experimental results on the Voice Conversion Challenge 2018 and 2020 (VCC2018 and VCC2020) datasets show that SVSNet outperforms well-known baseline systems in the assessment of speaker similarity at the utterance and system levels.
ISSN:2331-8422
DOI:10.48550/arxiv.2107.09392