Presentation 2007/7/19
Examination of criterion to distinguish between normal and unusual articulation
Akikazu NISHIKIDO, Jianwu DANG,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The current constraints do not suppress the problem of one-to-many in inverse estimation of articulation from speech sounds. One of the causes of the one-to-many problem is unusual articulation which can generate the same speech sound as that of normal articulation but has different articulatory manner and/or places from normal articulation. To exclude the unusual articulation from the inverse estimation, this study is going to find some criterion for distinguishing the normal and unusual articulations. For this purpose, we generated two types of articulatory and acoustic data set using a physiological articulatory model, and analyzed generated articulation for 5 Japanese vowels using principal component analysis. The results indicated that unusual articulation in the principal components space is locally located. Moreover, clustering for principal components was examined for distinguishing between normal articulation and unusual one.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Inverse Estimation / Unusual Articulation / Articulatory Model / Speech Production
Paper # SP2007-24
Date of Issue

Conference Information
Committee SP
Conference Date 2007/7/19(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) Examination of criterion to distinguish between normal and unusual articulation
Sub Title (in English)
Keyword(1) Inverse Estimation
Keyword(2) Unusual Articulation
Keyword(3) Articulatory Model
Keyword(4) Speech Production
1st Author's Name Akikazu NISHIKIDO
1st Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology()
2nd Author's Name Jianwu DANG
2nd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology
Date 2007/7/19
Paper # SP2007-24
Volume (vol) vol.107
Number (no) 165
Page pp.pp.-
#Pages 6
Date of Issue