000150578 001__ 150578
000150578 005__ 20190316234842.0
000150578 037__ $$aCONF
000150578 245__ $$aSparse Component Analysis for Speech Recognition in Multi-Speaker Environment
000150578 269__ $$a2010
000150578 260__ $$c2010
000150578 336__ $$aConference Papers
000150578 520__ $$aSparse Component Analysis is a relatively young technique that relies upon a representation of signal occupying only a small part of a larger space. Mixtures of sparse components are disjoint in that space. As a particular application of sparsity of speech signals, we investigate the DUET blind source separation algorithm in the context of speech recognition for multi-party recordings. We show how DUET can be tuned to the particular case of speech recognition with interfering sources, and evaluate the limits of performance as the number of sources increases. We show that the separated speech fits a common metric for sparsity, and conclude that sparsity assumptions lead to good performance in speech separation and hence ought to benefit other aspects of the speech recognition chain.
000150578 6531_ $$aAutomatic Speech Recognition
000150578 6531_ $$aOverlapping Speech
000150578 6531_ $$aSparse Component Analysis
000150578 700__ $$0243353$$g188259$$aAsaei, Afsaneh
000150578 700__ $$g117014$$aBourlard, Hervé$$0243348
000150578 700__ $$aGarner, Philip N.
000150578 7112_ $$cMakuhari, Japan$$aProceedings of Interspeech
000150578 8564_ $$uhttp://publications.idiap.ch/downloads/papers/2010/Asaei_INTERSPEECH_2010.pdf$$zURL
000150578 8564_ $$uhttps://infoscience.epfl.ch/record/150578/files/Asaei_INTERSPEECH_2010.pdf$$zn/a$$s581712
000150578 909C0 $$xU10381$$0252189$$pLIDIAP
000150578 909CO $$ooai:infoscience.tind.io:150578$$qGLOBAL_SET$$pconf$$pSTI
000150578 937__ $$aEPFL-CONF-150578
000150578 970__ $$aAsaei_INTERSPEECH_2010/LIDIAP
000150578 973__ $$sPUBLISHED$$aEPFL
000150578 980__ $$aCONF