Optimization of acoustic-to-articulatory mapping

Author

Prado, P.P.L. ; Shiva, E.H. ; Childers, D.G.

Author_Institution

Dept. of Electr. Eng., Florida Univ., FL, USA

Volume

2

fYear

1992

fDate

23-26 Mar 1992

Firstpage

33

Abstract

An optimization approach for estimating vocal tract configurations from the speech signal is presented. The scheme concatenates a gradient search and linear successive approximation, providing fast convergence and small errors. The objective function is generated by the least-absolute-value (l₁-norm) error between the model-derived and the speech-derived first three formants. The gradient search is accelerated by using an algorithm inspired by the Fletcher-Reeves method. Linear successive approximation assures convergence in the region near the optimum articulatory vector. Constraints are imposed on the articulatory parameters to avoid physiologically impossible configurations

Keywords

convergence; optimisation; speech analysis and processing; acoustic-to-articulatory mapping; constraints; fast convergence; gradient search; linear successive approximation; optimization; small errors; speech signal; vocal tract configurations; Acceleration; Convergence; Linear approximation; Measurement techniques; Multidimensional systems; Pharynx; Signal mapping; Speech analysis; Tongue; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on

Conference_Location

San Francisco, CA

ISSN

1520-6149

Print_ISBN

0-7803-0532-9

Type

conf

DOI

10.1109/ICASSP.1992.226127

Filename

226127