ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Fast acoustic likelihood computation using low-rank matrix approximation

Gajjar, Mrugesh R and Sreenivas, TV and Govindarajan, R (2011) Fast acoustic likelihood computation using low-rank matrix approximation. In: 2011 IEEE Workshop on Signal Processing Systems, October 4-7, 2011, Beirut, Lebanon.

[img] PDF
ieee_sips_322_2011.pdf - Published Version
Restricted to Registered users only

Download (202kB) | Request a copy
Official URL: http://www.informatik.uni-trier.de/~ley/db/conf/si...

Abstract

Acoustic modeling using mixtures of multivariate Gaussians is the prevalent approach for many speech processing problems. Computing likelihoods against a large set of Gaussians is required as a part of many speech processing systems and it is the computationally dominant phase for LVCSR systems. We express the likelihood computation as a multiplication of matrices representing augmented feature vectors and Gaussian parameters. The computational gain of this approach over traditional methods is by exploiting the structure of these matrices and efficient implementation of their multiplication.In particular, we explore direct low-rank approximation of the Gaussian parameter matrix and indirect derivation of low-rank factors of the Gaussian parameter matrix by optimum approximation of the likelihood matrix. We show that both the methods lead to similar speedups but the latter leads to far lesser impact on the recognition accuracy. Experiments on a 1138 word vocabulary RM1 task using Sphinx 3.7 system show that, for a typical case the matrix multiplication approach leads to overall speedup of 46%. Both the low-rank approximation methods increase the speedup to around 60%, with the former method increasing the word error rate (WER) from 3.2% to 6.6%, while the latter increases the WER from 3.2% to 3.5%.

Item Type: Conference Proceedings
Additional Information: Copyright of this article belongs to IEEE.
Department/Centre: Division of Electrical Sciences > Electrical Communication Engineering
Division of Interdisciplinary Research > Supercomputer Education & Research Centre
Depositing User: Id for Latest eprints
Date Deposited: 28 Mar 2013 06:14
Last Modified: 28 Mar 2013 06:15
URI: http://eprints.iisc.ac.in/id/eprint/46121

Actions (login required)

View Item View Item