Phonetic speaker recognition with support vector machines

DMCA

Phonetic speaker recognition with support vector machines

Cached

Download Links

Other Repositories/Bibliography

DBLP

by W. M. Campbell , J. P. Campbell , D. A. Reynolds , D. A. Jones , T. R. Leek

Venue:	in Advances in Neural Information Processing Systems
Citations:	47 - 6 self

BibTeX

@INPROCEEDINGS{Campbell_phoneticspeaker,
    author = {W. M. Campbell and J. P. Campbell and D. A. Reynolds and D. A. Jones and T. R. Leek},
    title = {Phonetic speaker recognition with support vector machines},
    booktitle = {in Advances in Neural Information Processing Systems},
    year = {},
    pages = {2004}
}

OpenURL

Abstract

A recent area of significant progress in speaker recognition is the use of high level features—idiolect, phonetic relations, prosody, discourse structure, etc. A speaker not only has a distinctive acoustic sound but uses language in a characteristic manner. Large corpora of speech data available in recent years allow experimentation with long term statistics of phone patterns, word patterns, etc. of an individual. We propose the use of support vector machines and term frequency analysis of phone sequences to model a given speaker. To this end, we explore techniques for text categorization applied to the problem. We derive a new kernel based upon a linearization of likelihood ratio scoring. We introduce a new phone-based SVM speaker recognition approach that halves the error rate of conventional phone-based approaches. 1

Keyphrases

support vector machine phonetic speaker recognition recent year conventional phone-based approach text categorization recent area word pattern speech data phonetic relation long term statistic likelihood ratio speaker recognition large corpus distinctive acoustic sound discourse structure phone sequence characteristic manner high level feature significant progress new kernel term frequency analysis error rate phone pattern

DMCA