Presentation 2001/5/18
Perceptual Evaluation of Naturalness Degradation Due to Substitution of Phonetic Environment for Concatenative Speech Synthesis
Hisashi KAWAI, Minoru TSUZAKI, Tsuyoshi MASUDA, Hideki IWASAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to estimate degradation of naturalness in concatenative speech synthesis due to mismatch of phonetic environments between unit selection and its use, perceptual experiments were conducted using speech stimuli synthesized by concatenating V and CV extracted from separate utterances. The results for substitution of the succeeding environment of a vowel showed that (1) the naturalness was low when a vowel segment was used before [y, w]; (2) environments of extraction and usage asymmetrically affected the naturalness. On the other hand, the results for substitution of the preceding environment of a consonant showed that the naturalhess was (1) high for vowel [i], while the naturalness was low for [N, a]; (2) affected more by the kind of consonant than by the preceding environment; (3) especially high when consonant was voiceless plosives, affricates, and fricatives.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) speech synthesis / waveform concatenation / unit selection / phonetic environment / naturalness / perception test
Paper # SP2001-22
Date of Issue

Conference Information
Committee SP
Conference Date 2001/5/18(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Perceptual Evaluation of Naturalness Degradation Due to Substitution of Phonetic Environment for Concatenative Speech Synthesis
Sub Title (in English)
Keyword(1) speech synthesis
Keyword(2) waveform concatenation
Keyword(3) unit selection
Keyword(4) phonetic environment
Keyword(5) naturalness
Keyword(6) perception test
1st Author's Name Hisashi KAWAI
1st Author's Affiliation ATR Spoken Language Translation Research Laboratories()
2nd Author's Name Minoru TSUZAKI
2nd Author's Affiliation ATR Spoken Language Translation Research Laboratories
3rd Author's Name Tsuyoshi MASUDA
3rd Author's Affiliation Nara Institute of Science and Technology
4th Author's Name Hideki IWASAWA
4th Author's Affiliation CREST
Date 2001/5/18
Paper # SP2001-22
Volume (vol) vol.101
Number (no) 87
Page pp.pp.-
#Pages 7
Date of Issue