A Comparison of Recent Waveform Generation and Acoustic Modeling Methods for Neural-Network-Based Speech Synthesis

Recent advances in speech synthesis suggest that limitations such as the lossy nature of the amplitude spectrum with minimum phase approximation and the over-smoothing effect in acoustic modeling can be overcome by using advanced machine learning approaches. In this paper, we build a framework in wh...

Full description

Saved in:
Bibliographic Details
Published in2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 4804 - 4808
Main Authors Wang, Xin, Lorenzo-Trueba, Jaime, Takaki, Shinji, Juvela, Lauri, Yamagishi, Junichi
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2018
Subjects
Acoustics
Artificial neural networks
autoregressive neural network
deep learning
Feature extraction
Gallium nitride
general adversarial network
Linguistics
Speech synthesis
Vocoders
Wavenet
Online AccessGet full text

Cover

More Information
Summary:Recent advances in speech synthesis suggest that limitations such as the lossy nature of the amplitude spectrum with minimum phase approximation and the over-smoothing effect in acoustic modeling can be overcome by using advanced machine learning approaches. In this paper, we build a framework in which we can fairly compare new vocoding and acoustic modeling techniques with conventional approaches by means of a large scale crowdsourced evaluation. Results on acoustic models showed that generative adversarial networks and an autoregressive (AR) model performed better than a normal recurrent network and the AR model performed best. Evaluation on vocoders by using the same AR acoustic model demonstrated that a Wavenet vocoder outperformed classical source-filter-based vocoders. Particularly, generated speech waveforms from the combination of AR acoustic model and Wavenet vocoder achieved a similar score of speech quality to vocoded speech.
ISSN:2379-190X
DOI:10.1109/ICASSP.2018.8461452