Loading [a11y]/accessibility-menu.js
An evaluation of a nonlinear feature transformation for conversational speech recognition | IEEE Conference Publication | IEEE Xplore

An evaluation of a nonlinear feature transformation for conversational speech recognition


Abstract:

We test the nonlinear symplectic maximum-likelihood transformation (SMLT) on two large-vocabulary, conversational speech recognition tasks: IBM's Superhuman test and the ...Show More

Abstract:

We test the nonlinear symplectic maximum-likelihood transformation (SMLT) on two large-vocabulary, conversational speech recognition tasks: IBM's Superhuman test and the DARPA 2003 Rich Transcription (RT03) test. Features in these tests are computed via linear discriminant analysis (LDA) on spliced MFCC features and subsequent transformation of the projected features using either a maximum-likelihood linear transformation (MLLT), an SMLT, or both. In contrast to previous tests of the SMLT on TIMIT phone recognition with static and delta MFCC, these tests use a more difficult task and very different features. The four results of this work are that both LDA+MLLT and LDA+SMLT systems outperform an LDA-only system; the LDA+MLLT system outperforms the LDA+SMLT system (but the MLLT has 20 times more parameters than the SMLT); small improvements over an LDA+MLLT system are obtained with an LDA+MLLT+SMLT system on well-matched material; and no improvements are obtained using two class-dependent SMLT in an LDA+MLLT+SMLT system.
Date of Conference: 17-21 May 2004
Date Added to IEEE Xplore: 30 August 2004
Print ISBN:0-7803-8484-9
Print ISSN: 1520-6149
Conference Location: Montreal, QC, Canada

References

References is not available for this document.