A fully-hardwired implementation of large vocabulary continuous speech recognizer

This article presents the hardware implementation of the speech recognition for real time performance and high-level accuracy. The stand-alone speech recognizer should simultaneously achieve the requirements, which are the low-latency performance and the low-power dissipation in an environment that...

Full description

Saved in:
Bibliographic Details
Main Authors: Yunjoo Kim, Juyeob Kim, Joohyun Lee, Wonjong Kim
Format: Conference Proceeding
Language:English
Subjects:
Consumer electronics
FPGA
Hidden Markov Model
Speech Recognition
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 2
container_issue
container_start_page 1
container_title
container_volume
creator Yunjoo Kim
Juyeob Kim
Joohyun Lee
Wonjong Kim
description This article presents the hardware implementation of the speech recognition for real time performance and high-level accuracy. The stand-alone speech recognizer should simultaneously achieve the requirements, which are the low-latency performance and the low-power dissipation in an environment that cannot connect to the network. So, we made a speech recognizer as the hardware accelerator based on the hidden Markov model for reducing the load of the system processor without the cloud computing. Our overall design has the fully hardwired operation flow from the generation of the speech feature to the generation of the recognized words. Our design showed low-latency performance as the real time factor of 0.4 ~ 0.5 on FPGA, which operates at 100MHz operating frequency and uses the resource of 10%.
doi_str_mv 10.1109/ISCE.2015.7177803
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_7177803</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>7177803</ieee_id><sourcerecordid>7177803</sourcerecordid><originalsourceid>FETCH-LOGICAL-i123t-40d1af444dc1e88e94ac3708b1b440bd88fca29887fefe149c1ceb882cb319a13</originalsourceid><addsrcrecordid>eNotkF1LwzAYRqMouE1_gHiTP9CaN0mb5HKUOQcDERW8G2n6Zov0i7RV5q934K6ec3U4PITcA0sBmHncvBWrlDPIUgVKaSYuyBxkroQSeZZfkhmHzCQgubgiM6akSvJcf96Q-TB8McaVEWZGXpfUT3V9TA42Vj8hYkVD09fYYDvaMXQt7Tytbdwj_e6cLacTH6nr2jG0UzcNdOgR3YFGdN2-Db8Yb8m1t_WAd-ddkI-n1XvxnGxf1ptiuU0CcDEmklVgvZSycoBao5HWCcV0CaWUrKy09s5yo7Xy6BGkceCw1Jq7UoCxIBbk4d8bEHHXx9CcynbnK8QfQktS0w</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>A fully-hardwired implementation of large vocabulary continuous speech recognizer</title><source>IEEE Xplore All Conference Series</source><creator>Yunjoo Kim ; Juyeob Kim ; Joohyun Lee ; Wonjong Kim</creator><creatorcontrib>Yunjoo Kim ; Juyeob Kim ; Joohyun Lee ; Wonjong Kim</creatorcontrib><description>This article presents the hardware implementation of the speech recognition for real time performance and high-level accuracy. The stand-alone speech recognizer should simultaneously achieve the requirements, which are the low-latency performance and the low-power dissipation in an environment that cannot connect to the network. So, we made a speech recognizer as the hardware accelerator based on the hidden Markov model for reducing the load of the system processor without the cloud computing. Our overall design has the fully hardwired operation flow from the generation of the speech feature to the generation of the recognized words. Our design showed low-latency performance as the real time factor of 0.4 ~ 0.5 on FPGA, which operates at 100MHz operating frequency and uses the resource of 10%.</description><identifier>ISSN: 0747-668X</identifier><identifier>EISSN: 2159-1423</identifier><identifier>EISBN: 1467373656</identifier><identifier>EISBN: 9781467373654</identifier><identifier>DOI: 10.1109/ISCE.2015.7177803</identifier><language>eng</language><publisher>IEEE</publisher><subject>Consumer electronics ; FPGA ; Hidden Markov Model ; Speech Recognition</subject><ispartof>2015 International Symposium on Consumer Electronics (ISCE), 2015, p.1-2</ispartof><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/7177803$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,777,781,786,787,23911,23912,25121,27906,54536,54913</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/7177803$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Yunjoo Kim</creatorcontrib><creatorcontrib>Juyeob Kim</creatorcontrib><creatorcontrib>Joohyun Lee</creatorcontrib><creatorcontrib>Wonjong Kim</creatorcontrib><title>A fully-hardwired implementation of large vocabulary continuous speech recognizer</title><title>2015 International Symposium on Consumer Electronics (ISCE)</title><addtitle>ISCE</addtitle><description>This article presents the hardware implementation of the speech recognition for real time performance and high-level accuracy. The stand-alone speech recognizer should simultaneously achieve the requirements, which are the low-latency performance and the low-power dissipation in an environment that cannot connect to the network. So, we made a speech recognizer as the hardware accelerator based on the hidden Markov model for reducing the load of the system processor without the cloud computing. Our overall design has the fully hardwired operation flow from the generation of the speech feature to the generation of the recognized words. Our design showed low-latency performance as the real time factor of 0.4 ~ 0.5 on FPGA, which operates at 100MHz operating frequency and uses the resource of 10%.</description><subject>Consumer electronics</subject><subject>FPGA</subject><subject>Hidden Markov Model</subject><subject>Speech Recognition</subject><issn>0747-668X</issn><issn>2159-1423</issn><isbn>1467373656</isbn><isbn>9781467373654</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2015</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotkF1LwzAYRqMouE1_gHiTP9CaN0mb5HKUOQcDERW8G2n6Zov0i7RV5q934K6ec3U4PITcA0sBmHncvBWrlDPIUgVKaSYuyBxkroQSeZZfkhmHzCQgubgiM6akSvJcf96Q-TB8McaVEWZGXpfUT3V9TA42Vj8hYkVD09fYYDvaMXQt7Tytbdwj_e6cLacTH6nr2jG0UzcNdOgR3YFGdN2-Db8Yb8m1t_WAd-ddkI-n1XvxnGxf1ptiuU0CcDEmklVgvZSycoBao5HWCcV0CaWUrKy09s5yo7Xy6BGkceCw1Jq7UoCxIBbk4d8bEHHXx9CcynbnK8QfQktS0w</recordid><startdate>20150601</startdate><enddate>20150601</enddate><creator>Yunjoo Kim</creator><creator>Juyeob Kim</creator><creator>Joohyun Lee</creator><creator>Wonjong Kim</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20150601</creationdate><title>A fully-hardwired implementation of large vocabulary continuous speech recognizer</title><author>Yunjoo Kim ; Juyeob Kim ; Joohyun Lee ; Wonjong Kim</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i123t-40d1af444dc1e88e94ac3708b1b440bd88fca29887fefe149c1ceb882cb319a13</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Consumer electronics</topic><topic>FPGA</topic><topic>Hidden Markov Model</topic><topic>Speech Recognition</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Yunjoo Kim</creatorcontrib><creatorcontrib>Juyeob Kim</creatorcontrib><creatorcontrib>Joohyun Lee</creatorcontrib><creatorcontrib>Wonjong Kim</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Yunjoo Kim</au><au>Juyeob Kim</au><au>Joohyun Lee</au><au>Wonjong Kim</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>A fully-hardwired implementation of large vocabulary continuous speech recognizer</atitle><btitle>2015 International Symposium on Consumer Electronics (ISCE)</btitle><stitle>ISCE</stitle><date>2015-06-01</date><risdate>2015</risdate><spage>1</spage><epage>2</epage><pages>1-2</pages><issn>0747-668X</issn><eissn>2159-1423</eissn><eisbn>1467373656</eisbn><eisbn>9781467373654</eisbn><abstract>This article presents the hardware implementation of the speech recognition for real time performance and high-level accuracy. The stand-alone speech recognizer should simultaneously achieve the requirements, which are the low-latency performance and the low-power dissipation in an environment that cannot connect to the network. So, we made a speech recognizer as the hardware accelerator based on the hidden Markov model for reducing the load of the system processor without the cloud computing. Our overall design has the fully hardwired operation flow from the generation of the speech feature to the generation of the recognized words. Our design showed low-latency performance as the real time factor of 0.4 ~ 0.5 on FPGA, which operates at 100MHz operating frequency and uses the resource of 10%.</abstract><pub>IEEE</pub><doi>10.1109/ISCE.2015.7177803</doi><tpages>2</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 0747-668X
ispartof 2015 International Symposium on Consumer Electronics (ISCE), 2015, p.1-2
issn 0747-668X
2159-1423
language eng
recordid cdi_ieee_primary_7177803
source IEEE Xplore All Conference Series
subjects Consumer electronics
FPGA
Hidden Markov Model
Speech Recognition
title A fully-hardwired implementation of large vocabulary continuous speech recognizer
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T13%3A48%3A32IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=A%20fully-hardwired%20implementation%20of%20large%20vocabulary%20continuous%20speech%20recognizer&rft.btitle=2015%20International%20Symposium%20on%20Consumer%20Electronics%20(ISCE)&rft.au=Yunjoo%20Kim&rft.date=2015-06-01&rft.spage=1&rft.epage=2&rft.pages=1-2&rft.issn=0747-668X&rft.eissn=2159-1423&rft_id=info:doi/10.1109/ISCE.2015.7177803&rft.eisbn=1467373656&rft.eisbn_list=9781467373654&rft_dat=%3Cieee_CHZPO%3E7177803%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i123t-40d1af444dc1e88e94ac3708b1b440bd88fca29887fefe149c1ceb882cb319a13%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=7177803&rfr_iscdi=true