Presentation 2010-12-10
Hardware speech recognition system for processing and recognition at moment
Masashi NAKAYAMA, Naoki SHIGEKAWA, Takashi YOKOUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we are proposed the hardware speech recognition system using FPGA (Field-Programmable Gate Array). Generally, conventional speech recognition systems are often constructed on computers and microcomputers as software and middleware. We thought that speech recognition systems will be constructed on hardware in future because developments of speech recognition systems change software to software. For these backgrounds, we are proposed]constructed the hardware speech recognition system with speech processing for the treatments which has speech signal processing part and speech recognition part. And, we confirmed the effectiveness of the system that the system is composed of the VAD (Voice Activity Detection) detector using summation of absolute difference between each signal in 2^ autocorrelation and 2-pass vowel speech recognition decoder using outputs of frequency band power and spectral distances. 2^ autocorrelation is calculated from 1^ autocorrelation with autocorrelation, and the decoder estimates the candidate using the frequency band power like filter bank and band-pass filter.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) FPGA / speech recognition / voice activity detection / noise environment / autocorrelation
Paper # EA2010-99
Date of Issue

Conference Information
Committee EA
Conference Date 2010/12/3(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 Engineering Acoustics (EA)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Hardware speech recognition system for processing and recognition at moment
Sub Title (in English)
Keyword(1) FPGA
Keyword(2) speech recognition
Keyword(3) voice activity detection
Keyword(4) noise environment
Keyword(5) autocorrelation
1st Author's Name Masashi NAKAYAMA
1st Author's Affiliation Department of Electrical and Computer Engineering, Kagawa National College of Technology()
2nd Author's Name Naoki SHIGEKAWA
2nd Author's Affiliation Faculty of Engineering, University of Fukui
3rd Author's Name Takashi YOKOUCHI
3rd Author's Affiliation Department of Electrical and Computer Engineering, Kagawa National College of Technology
Date 2010-12-10
Paper # EA2010-99
Volume (vol) vol.110
Number (no) 331
Page pp.pp.-
#Pages 6
Date of Issue