A solution to the reduction of concatenation artefacts in speech synthesis

Esther Klabbers; Raymond Veldhuis; Kim Koppen

A solution to the reduction of concatenation artefacts in speech synthesis

Esther Klabbers, Raymond Veldhuis, Kim Koppen

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

2 Downloads (Pure)

Abstract

One problem with speech synthesis impeding high quality is the occurrence of audible discontinuities at segment boundaries. Formant jumps across concatenation points suggest the problem to be due to spectral differences. The problem is most apparent in vowels and semi-vowels. We propose to reduce the number of audible discontinuities by adding context-sensitive diphones to the database. The number of additional diphones is limited by clustering contexts with similar spectral effects on the neighbouring vowels, using the Kullback-Leibler distance. A listening experiment has shown that the percentage of perceived discontinuities has significantly decreased.

Original language	English
Title of host publication	6th International Conference on Spoken Language Processing, ICSLP 2000
Place of Publication	Beijing
Publisher	China Military Friendship Pub.
Pages	474-477
Volume	3
ISBN (Print)	7801501144
Publication status	Published - 1 Jan 2000
Externally published	Yes
Event	6th International Conference on Spoken Language Processing, ICSLP 2000 - Beijing, China Duration: 16 Oct 2000 → 20 Oct 2000 Conference number: 6

Conference

Conference	6th International Conference on Spoken Language Processing, ICSLP 2000
Abbreviated title	ICSLP
Country	China
City	Beijing
Period	16/10/00 → 20/10/00

Access to Document

Klabbers2000solutionFinal published version, 72.1 KB

Fingerprint Dive into the research topics of 'A solution to the reduction of concatenation artefacts in speech synthesis'. Together they form a unique fingerprint.

Cite this

@inproceedings{2d8fdabfd6174704b79279bf3b68756a,

title = "A solution to the reduction of concatenation artefacts in speech synthesis",

abstract = "One problem with speech synthesis impeding high quality is the occurrence of audible discontinuities at segment boundaries. Formant jumps across concatenation points suggest the problem to be due to spectral differences. The problem is most apparent in vowels and semi-vowels. We propose to reduce the number of audible discontinuities by adding context-sensitive diphones to the database. The number of additional diphones is limited by clustering contexts with similar spectral effects on the neighbouring vowels, using the Kullback-Leibler distance. A listening experiment has shown that the percentage of perceived discontinuities has significantly decreased.",

author = "Esther Klabbers and Raymond Veldhuis and Kim Koppen",

year = "2000",

month = jan,

day = "1",

language = "English",

isbn = "7801501144",

volume = "3",

pages = "474--477",

booktitle = "6th International Conference on Spoken Language Processing, ICSLP 2000",

publisher = "China Military Friendship Pub.",

note = "6th International Conference on Spoken Language Processing, ICSLP 2000, ICSLP ; Conference date: 16-10-2000 Through 20-10-2000",

}

Klabbers, E, Veldhuis, R & Koppen, K 2000, A solution to the reduction of concatenation artefacts in speech synthesis. in 6th International Conference on Spoken Language Processing, ICSLP 2000. vol. 3, China Military Friendship Pub., Beijing, pp. 474-477, 6th International Conference on Spoken Language Processing, ICSLP 2000, Beijing, China, 16/10/00. <https://www.isca-speech.org/archive/icslp_2000/i00_3474.html>

A solution to the reduction of concatenation artefacts in speech synthesis. / Klabbers, Esther; Veldhuis, Raymond; Koppen, Kim.

6th International Conference on Spoken Language Processing, ICSLP 2000. Vol. 3 Beijing : China Military Friendship Pub., 2000. p. 474-477.

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

TY - GEN

T1 - A solution to the reduction of concatenation artefacts in speech synthesis

AU - Klabbers, Esther

AU - Veldhuis, Raymond

AU - Koppen, Kim

N1 - Conference code: 6

PY - 2000/1/1

Y1 - 2000/1/1

N2 - One problem with speech synthesis impeding high quality is the occurrence of audible discontinuities at segment boundaries. Formant jumps across concatenation points suggest the problem to be due to spectral differences. The problem is most apparent in vowels and semi-vowels. We propose to reduce the number of audible discontinuities by adding context-sensitive diphones to the database. The number of additional diphones is limited by clustering contexts with similar spectral effects on the neighbouring vowels, using the Kullback-Leibler distance. A listening experiment has shown that the percentage of perceived discontinuities has significantly decreased.

AB - One problem with speech synthesis impeding high quality is the occurrence of audible discontinuities at segment boundaries. Formant jumps across concatenation points suggest the problem to be due to spectral differences. The problem is most apparent in vowels and semi-vowels. We propose to reduce the number of audible discontinuities by adding context-sensitive diphones to the database. The number of additional diphones is limited by clustering contexts with similar spectral effects on the neighbouring vowels, using the Kullback-Leibler distance. A listening experiment has shown that the percentage of perceived discontinuities has significantly decreased.

UR - http://www.scopus.com/inward/record.url?scp=85009110388&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85009110388

SN - 7801501144

VL - 3

SP - 474

EP - 477

BT - 6th International Conference on Spoken Language Processing, ICSLP 2000

PB - China Military Friendship Pub.

CY - Beijing

T2 - 6th International Conference on Spoken Language Processing, ICSLP 2000

Y2 - 16 October 2000 through 20 October 2000

ER -