Research Articles (Electrical and Electronic Engineering)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/769

Browse

Browsing Research Articles (Electrical and Electronic Engineering) by Subject "Agglomerations, Industrial"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Clustering acoustic segments using multi- stage agglomerative hierarchical clustering
    (Public Library of Science, 2015) Lerato, Lerato; Niesler, Thomas
    Agglomerative hierarchical clustering becomes infeasible when applied to large datasets due to its O(N2) storage requirements. We present a multi-stage agglomerative hierarchical clustering (MAHC) approach aimed at large datasets of speech segments. The algorithm is based on an iterative divide-and-conquer strategy. The data is first split into independent subsets, each of which is clustered separately. Thus reduces the storage required for sequential implementations, and allows concurrent computation on parallel computing hardware. The resultant clusters are merged and subsequently re-divided into subsets, which are passed to the following iteration. We show that MAHC can match and even surpass the performance of the exact implementation when applied to datasets of speech segments.
  • Thumbnail Image
    Item
    Feature trajectory dynamic time warping for clustering of speech segments
    (SpringerOpen, 2019) Lerato, Lerato; Niesle, Thomas
    ENGLISH ABSTRACT: Dynamic time warping (DTW) can be used to compute the similarity between two sequences of generally differinglength. We propose a modification to DTW that performs individual and independent pairwise alignment of featuretrajectories. The modified technique, termed feature trajectory dynamic time warping (FTDTW), is applied as asimilarity measure in the agglomerative hierarchical clustering of speech segments. Experiments using MFCC and PLPparametrisations extracted from TIMIT and from the Spoken Arabic Digit Dataset (SADD) show consistent andstatistically significant improvements in the quality of the resulting clusters in terms of F-measure and normalisedmutual information (NMI).