Full Text Views
47
This article proposes a prosody correction method based on partial model adaptation for Chinese-accented Japanese hidden Markov model (HMM)-based text-to-speech synthesis. Although text-to-speech synthesis built from non-native speech accurately reproduces the speaker's individuality in synthetic speech, the naturalness of the synthetic speech is strongly degraded. In the proposed model, to improve the naturalness while preserving the speaker individuality of Chinese-accented Japanese text-to-speech synthesis, we partially utilize HMM parameters of native Japanese speech to synthesize prosody-corrected synthetic speech. Results of an experimental evaluation demonstrate that duration and F0 correction are significantly effective for improving naturalness.
Daiki SEKIZAWA
University of Tokyo
Shinnosuke TAKAMICHI
University of Tokyo
Hiroshi SARUWATARI
University of Tokyo
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Daiki SEKIZAWA, Shinnosuke TAKAMICHI, Hiroshi SARUWATARI, "Prosody Correction Preserving Speaker Individuality for Chinese-Accented Japanese HMM-Based Text-to-Speech Synthesis" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 6, pp. 1218-1221, June 2019, doi: 10.1587/transinf.2018EDL8264.
Abstract: This article proposes a prosody correction method based on partial model adaptation for Chinese-accented Japanese hidden Markov model (HMM)-based text-to-speech synthesis. Although text-to-speech synthesis built from non-native speech accurately reproduces the speaker's individuality in synthetic speech, the naturalness of the synthetic speech is strongly degraded. In the proposed model, to improve the naturalness while preserving the speaker individuality of Chinese-accented Japanese text-to-speech synthesis, we partially utilize HMM parameters of native Japanese speech to synthesize prosody-corrected synthetic speech. Results of an experimental evaluation demonstrate that duration and F0 correction are significantly effective for improving naturalness.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8264/_p
Copy
@ARTICLE{e102-d_6_1218,
author={Daiki SEKIZAWA, Shinnosuke TAKAMICHI, Hiroshi SARUWATARI, },
journal={IEICE TRANSACTIONS on Information},
title={Prosody Correction Preserving Speaker Individuality for Chinese-Accented Japanese HMM-Based Text-to-Speech Synthesis},
year={2019},
volume={E102-D},
number={6},
pages={1218-1221},
abstract={This article proposes a prosody correction method based on partial model adaptation for Chinese-accented Japanese hidden Markov model (HMM)-based text-to-speech synthesis. Although text-to-speech synthesis built from non-native speech accurately reproduces the speaker's individuality in synthetic speech, the naturalness of the synthetic speech is strongly degraded. In the proposed model, to improve the naturalness while preserving the speaker individuality of Chinese-accented Japanese text-to-speech synthesis, we partially utilize HMM parameters of native Japanese speech to synthesize prosody-corrected synthetic speech. Results of an experimental evaluation demonstrate that duration and F0 correction are significantly effective for improving naturalness.},
keywords={},
doi={10.1587/transinf.2018EDL8264},
ISSN={1745-1361},
month={June},}
Copy
TY - JOUR
TI - Prosody Correction Preserving Speaker Individuality for Chinese-Accented Japanese HMM-Based Text-to-Speech Synthesis
T2 - IEICE TRANSACTIONS on Information
SP - 1218
EP - 1221
AU - Daiki SEKIZAWA
AU - Shinnosuke TAKAMICHI
AU - Hiroshi SARUWATARI
PY - 2019
DO - 10.1587/transinf.2018EDL8264
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2019
AB - This article proposes a prosody correction method based on partial model adaptation for Chinese-accented Japanese hidden Markov model (HMM)-based text-to-speech synthesis. Although text-to-speech synthesis built from non-native speech accurately reproduces the speaker's individuality in synthetic speech, the naturalness of the synthetic speech is strongly degraded. In the proposed model, to improve the naturalness while preserving the speaker individuality of Chinese-accented Japanese text-to-speech synthesis, we partially utilize HMM parameters of native Japanese speech to synthesize prosody-corrected synthetic speech. Results of an experimental evaluation demonstrate that duration and F0 correction are significantly effective for improving naturalness.
ER -