Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation

Masatsune Tamura; Norbert Braunschweiler; Takehiko Kagoshima; Masami Akamine

doi:10.1109/ICASSP.2010.5495151

Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation

Masatsune Tamura, Norbert Braunschweiler, Takehiko Kagoshima, Masami Akamine

ENG - Graduate School of Engineering (Common)

Toshiba Corporation

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Citations (Scopus)

Abstract

In this paper, we propose a speech synthesis method that combines a natural waveform concatenation based speech synthesis method and our baseline plural unit selection and fusion method. Two main features of the proposed method are (i) prosody regeneration from selected speech units and (ii) using multiple speech units at non-adjacent segments. The non-adjacent segments is the segment that the previous or following speech units in the optimum speech unit sequence are not adjacent in the database. By using the prosody of selected speech units, the original prosodic expressions and sounds of recorded speech are retained, while discontinuities are reduced by using multiple speech units at non-adjacent segments. MOS evaluations showed that the proposed method provides a clear improvement against the conventional unit selection method and our baseline method.

Original language	English
Title of host publication	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	4802-4805
Number of pages	4
ISBN (Print)	9781424442966
DOIs	https://doi.org/10.1109/ICASSP.2010.5495151
Publication status	Published - 2010
Event	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States Duration: 2010 Mar 14 → 2010 Mar 19

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Conference

Conference	2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
Country/Territory	United States
City	Dallas, TX
Period	10/3/14 → 10/3/19

Keywords

Concatenative speech synthesis
Prosody generation
Unit fusion
Unit selection

Access to Document

10.1109/ICASSP.2010.5495151

Cite this

Tamura, M., Braunschweiler, N., Kagoshima, T., & Akamine, M. (2010). Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation. In 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings (pp. 4802-4805). Article 5495151 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2010.5495151

Tamura, Masatsune ; Braunschweiler, Norbert ; Kagoshima, Takehiko et al. / Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation. 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2010. pp. 4802-4805 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{9603fc9cd6ec4975b0d6b6209d2ab2db,

title = "Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation",

abstract = "In this paper, we propose a speech synthesis method that combines a natural waveform concatenation based speech synthesis method and our baseline plural unit selection and fusion method. Two main features of the proposed method are (i) prosody regeneration from selected speech units and (ii) using multiple speech units at non-adjacent segments. The non-adjacent segments is the segment that the previous or following speech units in the optimum speech unit sequence are not adjacent in the database. By using the prosody of selected speech units, the original prosodic expressions and sounds of recorded speech are retained, while discontinuities are reduced by using multiple speech units at non-adjacent segments. MOS evaluations showed that the proposed method provides a clear improvement against the conventional unit selection method and our baseline method.",

keywords = "Concatenative speech synthesis, Prosody generation, Unit fusion, Unit selection",

author = "Masatsune Tamura and Norbert Braunschweiler and Takehiko Kagoshima and Masami Akamine",

year = "2010",

doi = "10.1109/ICASSP.2010.5495151",

language = "English",

isbn = "9781424442966",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "4802--4805",

booktitle = "2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings",

address = "United States",

note = "2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 ; Conference date: 14-03-2010 Through 19-03-2010",

}

Tamura, M, Braunschweiler, N, Kagoshima, T & Akamine, M 2010, Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation. in 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings., 5495151, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 4802-4805, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 10/3/14. https://doi.org/10.1109/ICASSP.2010.5495151

Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation. / Tamura, Masatsune; Braunschweiler, Norbert; Kagoshima, Takehiko et al.
2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2010. p. 4802-4805 5495151 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation

AU - Tamura, Masatsune

AU - Braunschweiler, Norbert

AU - Kagoshima, Takehiko

AU - Akamine, Masami

PY - 2010

Y1 - 2010

N2 - In this paper, we propose a speech synthesis method that combines a natural waveform concatenation based speech synthesis method and our baseline plural unit selection and fusion method. Two main features of the proposed method are (i) prosody regeneration from selected speech units and (ii) using multiple speech units at non-adjacent segments. The non-adjacent segments is the segment that the previous or following speech units in the optimum speech unit sequence are not adjacent in the database. By using the prosody of selected speech units, the original prosodic expressions and sounds of recorded speech are retained, while discontinuities are reduced by using multiple speech units at non-adjacent segments. MOS evaluations showed that the proposed method provides a clear improvement against the conventional unit selection method and our baseline method.

AB - In this paper, we propose a speech synthesis method that combines a natural waveform concatenation based speech synthesis method and our baseline plural unit selection and fusion method. Two main features of the proposed method are (i) prosody regeneration from selected speech units and (ii) using multiple speech units at non-adjacent segments. The non-adjacent segments is the segment that the previous or following speech units in the optimum speech unit sequence are not adjacent in the database. By using the prosody of selected speech units, the original prosodic expressions and sounds of recorded speech are retained, while discontinuities are reduced by using multiple speech units at non-adjacent segments. MOS evaluations showed that the proposed method provides a clear improvement against the conventional unit selection method and our baseline method.

KW - Concatenative speech synthesis

KW - Prosody generation

KW - Unit fusion

KW - Unit selection

UR - http://www.scopus.com/inward/record.url?scp=78049384950&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049384950&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5495151

DO - 10.1109/ICASSP.2010.5495151

M3 - Conference contribution

AN - SCOPUS:78049384950

SN - 9781424442966

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 4802

EP - 4805

BT - 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010

Y2 - 14 March 2010 through 19 March 2010

ER -

Tamura M, Braunschweiler N, Kagoshima T, Akamine M. Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation. In 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2010. p. 4802-4805. 5495151. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2010.5495151

Unit selection speech synthesis using multiple speech units at non-adjacent segments for prosody and waveform generation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this