Comparison of clustering methods: A case study of text-independent speaker modeling

Abstract

Clustering is needed in various applications such as biometric person authentication, speech coding and recognition, image compression and information retrieval. Hundreds of clustering methods have been proposed for the task in various fields but, surprisingly, there are few extensive studies actually comparing them. An important question is how much the choice of a clustering method matters for the final pattern recognition application. Our goal is to provide a thorough experimental comparison of clustering methods for text-independent speaker verification. We consider parametric Gaussian mixture model (GMM) and non-parametric vector quantization (VQ) model using the best known clustering algorithms including iterative (K-means, random swap, expectation-maximization), hierarchical (pairwise nearest neighbor, split, split-and-merge), evolutionary (genetic algorithm), neural (self-organizing map) and fuzzy (fuzzy C-means) approaches. We study recognition accuracy, processing time, clustering validity, and correlation of clustering quality and recognition accuracy. Experiments from these complementary observations indicate clustering is not a critical task in speaker recognition and the choice of the algorithm should be based on computational complexity and simplicity of the implementation. This is mainly because of three reasons: the data is not clustered, large models are used and only the best algorithms are considered. For low-order models, choice of the algorithm, however, can have a significant effect. Index Terms

Performance quantification of clustering algorithms for false positive removal in fMRI by ROC curves

“…Furthermore, the k-means and SOM have dependence on the initial conditions, where the k-means method is strongly affected, and the SOM is influenced less (Kinnunen et al, 2011), as can be clearly seen in Figure 6, where the k-means deviation is higher than the SOM deviation. Therefore, when these algorithms are used, it is necessary to repeat the procedure several times (it was used 100 times in our experiment), and take a mean or median of the output.…”

On the Performance Degradation of Speaker Recognition System due to Variation in Speech Characteristics Caused by Physiological Changes

“…The collection of all code words is called a codebook. In [19], the authors compare different algorithms of codebook generation and conclude that the method of codebook generation is not so important. What is more important is the codebook size as it has a direct effect on the computational complexity.…”

A fuzzy-GMM classifier for multilingual speaker identification

“…Vectors from a large vector space can be mapped to a finite number of regions in the space known as clusters. Cluster analysis creates various clustering models to form an efficient speaker model.The best known models for speaker identification are centroid models that involve clustering algorithms like k-means algorithm, LBG algorithm, fuzzy c-means algorithm and expectation maximization algorithm [8].…”