Optimization of Perceptually-based ASR Frond-end
Identifieur interne : 000668 ( Crin/Curation ); précédent : 000667; suivant : 000669Optimization of Perceptually-based ASR Frond-end
Auteurs : H. Hermansky ; J.-C. JunquaSource :
English descriptors
- KwdEn :
Abstract
Several recently proposed automatic speech recognition (ASR) front-ends are experimentally compared for speaker-dependent and cross-speaker ASR. The perceptually based linear predictive (PLP) front-end yields the highest accuracies. By modifying its sensitivity to spectral peaks and to spectral tilt and by utilizing the speech dynamics we further improve by about 10========percnt; its error rate in speaker-independent ASR.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: Pour aller vers cette notice dans l'étape Curation :000668
Links to Exploration step
CRIN:hermansky88aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="185">Optimization of Perceptually-based ASR Frond-end</title></titleStmt><publicationStmt><idno type="RBID">CRIN:hermansky88a</idno><date when="1988" year="1988">1988</date><idno type="wicri:Area/Crin/Corpus">000668</idno><idno type="wicri:Area/Crin/Curation">000668</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">000668</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Optimization of Perceptually-based ASR Frond-end</title><author><name sortKey="Hermansky, H" sort="Hermansky, H" uniqKey="Hermansky H" first="H." last="Hermansky">H. Hermansky</name></author><author><name sortKey="Junqua, J C" sort="Junqua, J C" uniqKey="Junqua J" first="J.-C." last="Junqua">J.-C. Junqua</name></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>distance measure</term><term>front-end</term><term>speech recognition</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en" wicri:score="484">Several recently proposed automatic speech recognition (ASR) front-ends are experimentally compared for speaker-dependent and cross-speaker ASR. The perceptually based linear predictive (PLP) front-end yields the highest accuracies. By modifying its sensitivity to spectral peaks and to spectral tilt and by utilizing the speech dynamics we further improve by about 10========percnt; its error rate in speaker-independent ASR.</div></front></TEI><BibTex type="inproceedings"><ref>hermansky88a</ref><crinnumber>88-R-156</crinnumber><category>3</category><equipe>INCONNUE</equipe><author><e>Hermansky, H.</e><e>Junqua, J.-C.</e></author><title>Optimization of Perceptually-based ASR Frond-end</title><booktitle>{Proceedings ICASSP-88 (International Conference on Acoustic Speech and Signal Processing), New York (USA)}</booktitle><year>1988</year><pages>219-222</pages><month>apr</month><keywords><e>speech recognition</e><e>front-end</e><e>distance measure</e></keywords><abstract>Several recently proposed automatic speech recognition (ASR) front-ends are experimentally compared for speaker-dependent and cross-speaker ASR. The perceptually based linear predictive (PLP) front-end yields the highest accuracies. By modifying its sensitivity to spectral peaks and to spectral tilt and by utilizing the speech dynamics we further improve by about 10========percnt; its error rate in speaker-independent ASR.</abstract></BibTex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Crin/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000668 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Crin/Curation/biblio.hfd -nk 000668 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Crin
|étape= Curation
|type= RBID
|clé= CRIN:hermansky88a
|texte= Optimization of Perceptually-based ASR Frond-end
}}
| This area was generated with Dilib version V0.6.33. |  |