000203448 001__ 203448
000203448 005__ 20180913062846.0
000203448 0247_ $$2doi$$a10.1109/ICASSP.2010.5495662
000203448 037__ $$aCONF
000203448 245__ $$aSubspace Gaussian Mixture Models for speech recognition
000203448 269__ $$a2010
000203448 260__ $$c2010
000203448 336__ $$aConference Papers
000203448 520__ $$aWe describe an acoustic modeling approach in which all phonetic states share a common Gaussian Mixture Model structure, and the means and mixture weights vary in a subspace of the total parameter space. We call this a Subspace Gaussian Mixture Model (SGMM). Globally shared parameters define the subspace. This style of acoustic model allows for a much more compact representation and gives better results than a conventional modeling approach, particularly with smaller amounts of training data.
000203448 6531_ $$aSpeech Recognition
000203448 6531_ $$aHidden Markov Models
000203448 6531_ $$aGaussian Mixture Models
000203448 700__ $$aPovey, Daniel
000203448 700__ $$aBurget, Lukas
000203448 700__ $$aAgarwal, Mohit
000203448 700__ $$0248490$$aAkyazi, Pinar$$g238998
000203448 700__ $$aFeng, Kai
000203448 700__ $$aGhoshal, Arnab
000203448 700__ $$aGlembek, Ondrej
000203448 700__ $$aGoel, Nagendra Kumar
000203448 700__ $$aKarafiat, Martin
000203448 700__ $$aRastrow, Ariya
000203448 700__ $$aRose, Richard C.
000203448 700__ $$aSchwarz, Petr
000203448 700__ $$aThomas, Samuel
000203448 7112_ $$aIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)$$d2010
000203448 773__ $$q4330-4333$$t2010 IEEE International Conference on Acoustics, Speech and Signal Processing
000203448 909C0 $$0252445$$pIEL$$xU10318
000203448 909CO $$ooai:infoscience.tind.io:203448$$pconf$$pSTI
000203448 917Z8 $$x238998
000203448 917Z8 $$x148230
000203448 937__ $$aEPFL-CONF-203448
000203448 973__ $$aOTHER$$rREVIEWED$$sPUBLISHED
000203448 980__ $$aCONF