1999/03/01 by Petros Maragos, Alexandros Potamianos · 125 citations
Computer Science · Mathematics · #Music and Audio Processing #Speech and Audio Processing #Speech Recognition and Synthesis #Fractal dimension #Speech recognition #Computer science #Fractal #Hidden Markov model #Computation #Segmentation #Speech processing #Dimension (graph theory) #Speech segmentation #Pattern recognition (psychology) #Artificial intelligence #Mathematics #Algorithm
paper · open access · doi:10.1121/1.426738
published in The Journal of the Acoustical Society of America 105(3), 1925-1932 (Acoustical Society of America)
openalex publication_date 1999/03/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/16
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.