STiLL - Speech Technology in Language Learning

May 25-27, 1998
Marholmen, Sweden

A Signal Processing Technique for Speech Visualization

A. Álvarez, R. Martinez, P. Gómez, J. L. Domínguez

Departamento de Arquitectura y Tecnologia de Sistemas Informáticos Universidad PolitScnica de Madrid, Madrid, Spain

Through the present paper, a methodology to create Visual Representations of Speech for Speech Perception Enhancement Applications is presented. The specific mathematical and computational issues, based on the use of Gradient-Adaptive Lattices, Energy-Band Extraction and a Continuous Formant-Tracking Algorithm, are given. A specific case for Computer-Aided Language Learning oriented to the Phonetic Specificities of English for Spanish Speakers is also presented. The usual inconvenience of adapting the system to each specific speaker may be relaxed using a proper normalization method.

Full Paper

Bibliographic reference.  Álvarez, A. / Martinez, R. / Gómez, P. / Domínguez, J. L. (1998): "A signal processing technique for speech visualization", In STiLL-1998, 33-36.