Presentation 1995/12/15
Word spotting using a contextual effect model
Yuji Yonezawa, Masato Akagi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Coarticulation reduces word spotting accuracy in continuous speech when isolated uttered words are used as reference patterns for continuous DTW. In this paper, a contextual effect model was used as a front-end processor for word spotting to cope with coarticulation effects. The model adopted perceptual shifts by adjacent spectral peaks and has the ability to recover neutralized phonemes in continuous speech and can reduce vowel recognition error rates. The experiment results show that the model can improve word spotting accuracy.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) word spotting / continuous dynamic programming / contextual effect / coarticulation
Paper # SP95-108
Date of Issue

Conference Information
Committee SP
Conference Date 1995/12/15(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) Word spotting using a contextual effect model
Sub Title (in English)
Keyword(1) word spotting
Keyword(2) continuous dynamic programming
Keyword(3) contextual effect
Keyword(4) coarticulation
1st Author's Name Yuji Yonezawa
1st Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology()
2nd Author's Name Masato Akagi
2nd Author's Affiliation School of Information Science, Japan Advanced Institute of Science and Technology
Date 1995/12/15
Paper # SP95-108
Volume (vol) vol.95
Number (no) 431
Page pp.pp.-
#Pages 6
Date of Issue