Clustering of Cepstrum Coefficients Using Pairwise Mutual Information
نویسنده
چکیده
In this paper I consider the problem of clustering the cepstrum coefficients of an acoustic vector into a number of disjoint sets (subvectors) using the mutual information as the clustering criterion. I then quantize each one of the subvectors independently using different quantization step. I compare the performance of the clustering scheme with a heuristic one where neighboring coefficients are clustered together. Preliminary results show that the proposed clustering schemes do not improve the Word Error Rate (WER) given a fixed total number of bits.
منابع مشابه
Clustering of a Number of Genes Affecting in Milk Production using Information Theory and Mutual Information
Information theory is a branch of mathematics. Information theory is used in genetic and bioinformatics analyses and can be used for many analyses related to the biological structures and sequences. Bio-computational grouping of genes facilitates genetic analysis, sequencing and structural-based analyses. In this study, after retrieving gene and exon DNA sequences affecting milk yield in dairy ...
متن کاملPairwise clustering based on the mutual-information criterion
Pairwise clustering methods partition a dataset using pairwise similarity between data-points. The pairwise similarity matrix can be used to define a Markov random walk on the data points. This view forms a probabilistic interpretation of spectral clustering methods. We utilize this probabilistic model to define a novel clustering cost function that is based on maximizing the mutual information...
متن کاملAn evaluation of using mutual information for selection of acoustic-features representation of phonemes for speech recognition
This paper addresses the problem of finding a subset of the acoustic feature space that best represents the phoneme set used in a speech recognition system. A maximum mutual information approach is presented for selecting acoustic features to be combined together to represent the distinctions among the phonemes. The overall phoneme recognition accuracy is slightly increased for the same length ...
متن کاملInformation Theoretic Pairwise Clustering
In this paper we develop an information-theoretic approach for pairwise clustering. The Laplacian of the pairwise similarity matrix can be used to define a Markov random walk on the data points. This view forms a probabilistic interpretation of spectral clustering methods. We utilize this probabilistic model to define a novel clustering cost function that is based on maximizing the mutual infor...
متن کاملMaximum Likelihood and Maximum Mutual Information Training in Gender and Age Recognition System
Gender and age estimation based on Gaussian Mixture Models (GMM) is introduced. Telephone recordings from the Czech SpeechDatEast database are used as training and test data set. Mel-Frequency Cepstral Coefficients (MFCC) are extracted from the speech recordings. To estimate the GMMs’ parameters Maximum Likelihood (ML) training is applied. Consequently these estimations are used as the baseline...
متن کامل