Presentation 1995/11/16
A word recognition using demi-syllable wordspotting
Masaru TAKANO, Ken-ichi ISO, Takao WATANABE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) For robust word recognition , it is important to deal with utterances including a keyword and extra non-keywords. In this report, we present a novel word spotting method using likelihood normalization and inhibition for overlapped word detection. The method does not require speech detection which is difficult for utterances with extra non-keywords and it reduces keyword insertion errors using the inhibition for overlapped word detection. We realize the frame synchronous algorithm for in for overlapped word detection using word internal likelihood as a cue for overlap. It is experimentally shown that the proposed method works well.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) word recognition / non-keyword / robustness / wordspotting / likelihood normalization, inhibition for overlapped word detection
Paper # SP95-77
Date of Issue

Conference Information
Committee SP
Conference Date 1995/11/16(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) A word recognition using demi-syllable wordspotting
Sub Title (in English)
Keyword(1) word recognition
Keyword(2) non-keyword
Keyword(3) robustness
Keyword(4) wordspotting
Keyword(5) likelihood normalization, inhibition for overlapped word detection
1st Author's Name Masaru TAKANO
1st Author's Affiliation NEC Information Technology Research Laboratories()
2nd Author's Name Ken-ichi ISO
2nd Author's Affiliation NEC Information Technology Research Laboratories
3rd Author's Name Takao WATANABE
3rd Author's Affiliation NEC Information Technology Research Laboratories
Date 1995/11/16
Paper # SP95-77
Volume (vol) vol.95
Number (no) 355
Page pp.pp.-
#Pages 8
Date of Issue