Premium
Pole–zero estimation of speech signal based on zero‐tracking algorithm
Author(s) -
Ouaaline N.,
Radouane L.
Publication year - 1998
Publication title -
international journal of adaptive control and signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.73
H-Index - 66
eISSN - 1099-1115
pISSN - 0890-6327
DOI - 10.1002/(sici)1099-1115(199802)12:1<1::aid-acs439>3.0.co;2-a
Subject(s) - formant , zero (linguistics) , autoregressive model , algorithm , signal (programming language) , context (archaeology) , extension (predicate logic) , tracking (education) , autoregressive–moving average model , speech recognition , computer science , adaptive algorithm , speech processing , mathematics , statistics , vowel , psychology , paleontology , pedagogy , linguistics , philosophy , biology , programming language
This paper describes an adaptive algorithm for formant and antiformant tracking of nasal sounds of speech signal. It is based on an extension of the zero‐tracking algorithm (ZTA) to the autoregressive moving average model (ARMA). In this context the parameter estimation algorithm used is a modified RLS algorithm (σ‐RLS). This adaptive algorithm is shown to be robust against the sounds. The proposed algorithm updates adaptively the complex poles (formants) of the AR part and the complex zeros (antiformants) of the MA part respectively. An application to natural speech is given. © 1998 John Wiley & Sons, Ltd.