Optimal clustering of multivariate normal distributions using divergence and its application to HMM adaptation

Myrvoll, Tor André; Soong, Frank K.

doi:10.1109/icassp.2003.1198840

Cited by 22 publications

(22 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The centroid is the CCG that minimizes the sum of the WSKLD to all CGs. Here, we extend the results of [43] by modifying the cost function to (21). The mean of a CCG is thereby estimated aŝ…”

Section: Parameter Estimation Of Ccgsmentioning

confidence: 99%

“…Like in [43],ˆ n is constrained to be diagonal during clustering. It can be seen from Equations (22) and (23) that the procedure of estimating the CCGs given the weights ĝ(n,m) is iterative.…”

Section: Parameter Estimation Of Ccgsmentioning

confidence: 99%

“…The estimation of means and covariance matrices of the CCGs is based on the approach in [43], where a method for finding the centroid of a set of Gaussians is derived. The centroid is the CCG that minimizes the sum of the WSKLD to all CGs.…”

Section: Parameter Estimation Of Ccgsmentioning

confidence: 99%

See 2 more Smart Citations

Multi-candidate missing data imputation for robust speech recognition

Wang

hamme

2012

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

The application of Missing Data Techniques (MDT) to increase the noise robustness of HMM/GMM-based large vocabulary speech recognizers is hampered by a large computational burden. The likelihood evaluations imply solving many constrained least squares (CLSQ) optimization problems. As an alternative, researchers have proposed frontend MDT or have made oversimplifying independence assumptions for the backend acoustic model. In this article, we propose a fast Multi-Candidate (MC) approach that solves the per-Gaussian CLSQ problems approximately by selecting the best from a small set of candidate solutions, which are generated as the MDT solutions on a reduced set of cluster Gaussians. Experiments show that the MC MDT runs equally fast as the uncompensated recognizer while achieving the accuracy of the full backend optimization approach. The experiments also show that exploiting the more accurate acoustic model of the backend does pay off in terms of accuracy when compared to frontend MDT.

show abstract

Section: Parameter Estimation Of Ccgsmentioning

confidence: 99%

Section: Parameter Estimation Of Ccgsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-candidate missing data imputation for robust speech recognition

Wang

hamme

2012

J AUDIO SPEECH MUSIC PROC.

View full text Add to dashboard Cite

show abstract

“…Clustering "raw" normal data sets is also an important algorithmic issue in computer vision and sound processing. For example, Myrvoll and Soong [3] consider this task for adapting hidden Markov model (HMM) parameters in a structured maximum a posteriori linear regression (SMAPLR), and obtained improved speech recognition rate. In computer vision, Gaussian mixture models (GMMs) abound from statistical image modeling learnt by the expectation-maximization (EM) soft clustering technique [4], and therefore represent a versatile source of raw Gaussian data sets to manipulate efficiently.…”

Section: Introductionmentioning

confidence: 99%

“…This is all the more important for applications that require to handle symmetric information-theoretic measures [3].…”

mentioning

confidence: 99%

Clustering Multivariate Normal Distributions

Nielsen

Nock²

2009

Emerging Trends in Visual Computing

View full text Add to dashboard Cite

Abstract. In this paper, we consider the task of clustering multivariate normal distributions with respect to the relative entropy into a prescribed number, k, of clusters using a generalization of Lloyd's k-means algorithm [1]. We revisit this information-theoretic clustering problem under the auspices of mixed-type Bregman divergences, and show that the approach of Davis and Dhillon [2] (NIPS*06) can also be derived directly, by applying the Bregman k-means algorithm, once the proper vector/matrix Legendre transformations are defined. We further explain the dualistic structure of the sided k-means clustering, and present a novel k-means algorithm for clustering with respect to the symmetrical relative entropy, the J-divergence. Our approach extends to differential entropic clustering of arbitrary members of the same exponential families in statistics.

show abstract

Automatic Speech Recognition on Mobile Devices and over Communication Networks

Tan

Lindberg

2008

Advances in Pattern Recognition

View full text Add to dashboard Cite

Optimal clustering of multivariate normal distributions using divergence and its application to HMM adaptation

Cited by 22 publications

References 11 publications

Multi-candidate missing data imputation for robust speech recognition

Multi-candidate missing data imputation for robust speech recognition

Clustering Multivariate Normal Distributions

Automatic Speech Recognition on Mobile Devices and over Communication Networks

Contact Info

Product

Resources

About