5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

On Using Fractal Features of Speech Sounds in Automatic Speech Recognition

Petros Maragos (1), Alexandros Potamianos (2)

(1) Institute for Language & Speech Processing, Athens, Greece School of E.C.E., Georgia Institute of Technology, Atlanta, GA, USA (2) AT&T Labs-Research, Florham Park, NJ, USA

The dynamics of air ow during speech production may often result into some small or large degree of turbulence. In this paper, we quantify the geometry of speech turbulence as reflected in the fragmentation of the time signal by using fractal models. We describe an efficient algorithm for estimating the short-time fractal dimension of speech signals based on multiscale morphological filtering and discuss its potential for phonetic classification. We also report 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 provides a modest improvement in speech recognition performance. dimensions of speech segments as additional features in an automatic speech recognition system based on hidden Markov models (HMMs) and found them to offer a modest improvement to the speech recognition performance.

Full Paper

Bibliographic reference.  Maragos, Petros / Potamianos, Alexandros (1997): "On using fractal features of speech sounds in automatic speech recognition", In EUROSPEECH-1997, 2531-2534.