This paper gives a general overview of hidden Markov model (HMM)-based speech synthesis, which has recently been demonstrated to be very effective in synthesizing speech. The main advantage of this approach is its flexibility in changing speaker identities, emotions, and speaking styles. This paper also discusses the relation between the HMM-based approach and the more conventional unit-selection approach that has dominated over the last decades. Finally, advanced techniques for future developments are described.
%0 Journal Article
%1 Tokuda2013
%A Tokuda, Keiichi
%A Nankaku, Yoshihiko
%A Toda, Tomoki
%A Zen, Heiga
%A Yamagishi, Junichi
%A Oura, Keiichiro
%D 2013
%J Proceedings of the IEEE
%K (HMM);statistical (TTS) Markov approach;Hidden hidden learning;Text model model;speech models;Information models;speech parametric processing;HMM-based processing;Parametric processing;Speech speech statistics;Speech synthesis synthesis;Statistical synthesis;hidden synthesis;text-to-speech synthesis;unit-selection system;Hidden
%N 5
%P 1234-1252
%R 10.1109/JPROC.2013.2251852
%T Speech Synthesis Based on Hidden Markov Models
%V 101
%X This paper gives a general overview of hidden Markov model (HMM)-based speech synthesis, which has recently been demonstrated to be very effective in synthesizing speech. The main advantage of this approach is its flexibility in changing speaker identities, emotions, and speaking styles. This paper also discusses the relation between the HMM-based approach and the more conventional unit-selection approach that has dominated over the last decades. Finally, advanced techniques for future developments are described.
@article{Tokuda2013,
abstract = {This paper gives a general overview of hidden Markov model (HMM)-based speech synthesis, which has recently been demonstrated to be very effective in synthesizing speech. The main advantage of this approach is its flexibility in changing speaker identities, emotions, and speaking styles. This paper also discusses the relation between the HMM-based approach and the more conventional unit-selection approach that has dominated over the last decades. Finally, advanced techniques for future developments are described.},
added-at = {2021-02-01T10:51:23.000+0100},
author = {Tokuda, Keiichi and Nankaku, Yoshihiko and Toda, Tomoki and Zen, Heiga and Yamagishi, Junichi and Oura, Keiichiro},
biburl = {https://www.bibsonomy.org/bibtex/22718b2a4c86e019335e73f02b4f9a378/m-toman},
doi = {10.1109/JPROC.2013.2251852},
file = {:pdfs/tokuda_procieee_2013.pdf:PDF},
interhash = {c439bda2b37d89a03d42c1e8944a0eb4},
intrahash = {2718b2a4c86e019335e73f02b4f9a378},
issn = {0018-9219},
journal = {Proceedings of the IEEE},
keywords = {(HMM);statistical (TTS) Markov approach;Hidden hidden learning;Text model model;speech models;Information models;speech parametric processing;HMM-based processing;Parametric processing;Speech speech statistics;Speech synthesis synthesis;Statistical synthesis;hidden synthesis;text-to-speech synthesis;unit-selection system;Hidden},
month = apr,
number = 5,
owner = {schabus},
pages = {1234-1252},
timestamp = {2021-02-01T10:51:23.000+0100},
title = {Speech Synthesis Based on Hidden Markov Models},
volume = 101,
year = 2013
}