Fractal dimensions of speech sounds: Computation and application to automatic speech recognition

Maragos, P; Potamianos, A

dc.contributor.author	Maragos, P	en
dc.contributor.author	Potamianos, A	en
dc.date.accessioned	2014-03-01T01:14:41Z
dc.date.available	2014-03-01T01:14:41Z
dc.date.issued	1999	en
dc.identifier.issn	0001-4966	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/13180
dc.subject	Automatic Speech Recognition	en
dc.subject	Fractal Dimension	en
dc.subject.classification	Acoustics	en
dc.subject.other	airflow	en
dc.subject.other	algorithm	en
dc.subject.other	article	en
dc.subject.other	mathematical analysis	en
dc.subject.other	model	en
dc.subject.other	priority journal	en
dc.subject.other	research	en
dc.subject.other	signal processing	en
dc.subject.other	spectroscopy	en
dc.subject.other	speech analysis	en
dc.subject.other	speech discrimination	en
dc.subject.other	waveform	en
dc.subject.other	word recognition	en
dc.subject.other	Automatic Data Processing	en
dc.subject.other	Databases	en
dc.subject.other	Fractals	en
dc.subject.other	Humans	en
dc.subject.other	Markov Chains	en
dc.subject.other	Models, Biological	en
dc.subject.other	Phonetics	en
dc.subject.other	Sound	en
dc.subject.other	Speech	en
dc.subject.other	Speech Perception	en
dc.subject.other	Speech Production Measurement	en
dc.subject.other	Time Factors	en
dc.title	Fractal dimensions of speech sounds: Computation and application to automatic speech recognition	en
heal.type	journalArticle	en
heal.identifier.primary	10.1121/1.426738	en
heal.identifier.secondary	http://dx.doi.org/10.1121/1.426738	en
heal.language	English	en
heal.publicationDate	1999	en
heal.abstract	The dynamics of airflow during speech production may often result in some small or large degree of turbulence. In this paper, the geometry of speech turbulence as reflected in the fragmentation of the time signal is quantified by using fractal models. An efficient algorithm for estimating the short-time fractal dimension of speech signals based on multiscale morphological filtering is described, and its potential for speech segmentation and phonetic classification discussed. Also reported are experimental results on using the short-time fractal dimension of speech signals at multiple scales as additional features in an automatic speech-recognition system using hidden Markov models, which provide a modest improvement in speech-recognition performance. (C) 1999 Acoustical Society of America. [S0001-4966(99)02403-0].	en
heal.publisher	AMER INST PHYSICS	en
heal.journalName	Journal of the Acoustical Society of America	en
dc.identifier.doi	10.1121/1.426738	en
dc.identifier.isi	ISI:000079078500049	en
dc.identifier.volume	105	en
dc.identifier.issue	3	en
dc.identifier.spage	1925	en
dc.identifier.epage	1932	en