Presentation 2003/1/24
Auditory Evaluation of Japanese Vowel Synthesized by Using Cascaded Sand-glass Type Neural Network
Masaya KIMOTO, Tadaaki SHIMIZU, Hiroki YOSHIMURA, Naoki ISU, Kazuhiro SUGATA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We proposed a new scheme to derive the characteristics of vowels from LSP parameters as well as compress information by using cascaded sand-glass type neural network (nonlinear, 5-layers) (CSNN(NL5)). Here, it was shown that LSP parameters can be restored with well feasibility for practical use of speech synthesis from the compressed LSP parameters. In order to verify their audibility, we performed auditory evaluation of synthesized speech by using artificial parameters on the hidden layer output plane. As a result, we verified the compressed LSP parameters by our scheme have workable quality for speech synthesis.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Sand-glass Type Neural Network / LSP Analysis / Vowel / Formant / Principal Component Analysis
Paper #
Date of Issue

Conference Information
Committee SP
Conference Date 2003/1/24(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) Auditory Evaluation of Japanese Vowel Synthesized by Using Cascaded Sand-glass Type Neural Network
Sub Title (in English)
Keyword(1) Sand-glass Type Neural Network
Keyword(2) LSP Analysis
Keyword(3) Vowel
Keyword(4) Formant
Keyword(5) Principal Component Analysis
1st Author's Name Masaya KIMOTO
1st Author's Affiliation Tottori University, Faculty of Engineering, Department of Information and Knowledge Engineering()
2nd Author's Name Tadaaki SHIMIZU
2nd Author's Affiliation Tottori University, Faculty of Engineering, Department of Information and Knowledge Engineering
3rd Author's Name Hiroki YOSHIMURA
3rd Author's Affiliation Tottori University, Faculty of Engineering, Department of Information and Knowledge Engineering
4th Author's Name Naoki ISU
4th Author's Affiliation Tottori University, Faculty of Engineering, Department of Information and Knowledge Engineering
5th Author's Name Kazuhiro SUGATA
5th Author's Affiliation Tottori University, Faculty of Engineering, Department of Information and Knowledge Engineering
Date 2003/1/24
Paper #
Volume (vol) vol.102
Number (no) 619
Page pp.pp.-
#Pages 6
Date of Issue