Presentation 2007/12/13
Dynamic Integration of Multiple Feature Streams Utilizing Entropy of Search Hypotheses
Shoei SATO, Takahiro OKU, Shinich HOMMA, Akio KOBAYASHI, Toru IMAI, Tohru TAKAGI, Tetsunori KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We present a novel method of integrating the likelihoods of multiple feature streams, representing different acoustic aspects, for robust speech recognition. The proposed integration algorithm dynamically calculates a frame-wise stream weight based on an entropy reduction caused by observing an input feature. Japanese broadcast news recognition experiments, integrating feature streams representing energy, amplitude drifts, and resonant frequency drifts, showed that the proposed method reduced error words by 9.2% in field reports and 4.7% in spontaneous commentaries relative to the best result obtained from a single stream.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Speech recognition / Stream integration / Entropy
Paper # NLC2007-39,SP2007-102
Date of Issue

Conference Information
Committee NLC
Conference Date 2007/12/13(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 Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Dynamic Integration of Multiple Feature Streams Utilizing Entropy of Search Hypotheses
Sub Title (in English)
Keyword(1) Speech recognition
Keyword(2) Stream integration
Keyword(3) Entropy
1st Author's Name Shoei SATO
1st Author's Affiliation NHK Science and Technical Research Laboratories:Faculty of Science and Engineering, Waseda University()
2nd Author's Name Takahiro OKU
2nd Author's Affiliation NHK Science and Technical Research Laboratories
3rd Author's Name Shinich HOMMA
3rd Author's Affiliation NHK Science and Technical Research Laboratories
4th Author's Name Akio KOBAYASHI
4th Author's Affiliation NHK Science and Technical Research Laboratories
5th Author's Name Toru IMAI
5th Author's Affiliation NHK Science and Technical Research Laboratories
6th Author's Name Tohru TAKAGI
6th Author's Affiliation NHK Science and Technical Research Laboratories
7th Author's Name Tetsunori KOBAYASHI
7th Author's Affiliation Faculty of Science and Engineering, Waseda University
Date 2007/12/13
Paper # NLC2007-39,SP2007-102
Volume (vol) vol.107
Number (no) 405
Page pp.pp.-
#Pages 6
Date of Issue