Tecnología de voz utilizada en la terapia del lenguaje de niños con deficiencias auditivas.

ABSTRACT Resumen. En este documento se presenta la investigación, análisis y diseño de un módulo de enseñanza de lenguaje oral para niños con deficiencias auditivas. En el módulo se propone el método de aprender el lenguaje oral escuchando, leyendo y pronunciando palabras y/o frases comunes del español, apoyado en un tutor animado que muestra el lugar y forma de articulación, imágenes y sonidos digitalizados. La herramienta se desarrolla utilizando estrategias de educación especial y tecnología de reconocimiento de voz. Dicho módulo será el primero en formar parte de un sistema denominado SAEL: Sistema de Apoyo a la Enseñanza del Lenguaje, proyecto que apoya la enseñanza del lenguaje bilingüe (de señas y oral) de los niños del Centro de Atención Múltiple de Cancún, México. Abstrac. This paper presents the research, analysis and design of the Language Teaching Module for the SAEL System: Teaching of the Language Support System. This project supports bilingual language teaching (sign and oral) to hearing impaired children using strategies and methods in step with the needs of schools and teachers of Special Education. This module proposes a method designed to teach oral language reading and pronouncing words and / or phrases that are commonly used in Spanish with the help of an enthusiastic tutor who shows how to articulate the words or expressions. The method works by using methods in special educational strategies applying voice recognition and digital image technologies for the teaching and practice of basic language elements.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This work is part of an effort aimed at developing computer-based systems for language instruction; we address the task of grading the pronunciation quality of the speech of a student of a foreign language. The automatic grading system uses SRI's Decipher<sup>TM</sup> continuous speech recognition system to generate phonetic segmentations. Based on these segmentations and probabilistic models we produce pronunciation scores for individual or groups of sentences. Scores obtained from expert human listeners are used as the reference to evaluate the different machine scores and to provide targets when training some of the algorithms. In previous work we had found that duration-based scores outperformed HMM log-likelihood-based scores. In this paper we show that we can significantly improve HMM-based scores by using average phone segment posterior probabilities. Correlation between machine and human scores went up from r=0.50 with likelihood-based scores to r=0.88 with posterior-based scores. The new measures also outperformed duration-based scores in their ability to produce reliable scores from only a few sentences
    Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on; 05/1997
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A growing assemblage of researchers has, in recent years, adopted methods and theories that acknowledge and exploit the multisensory nature of speech perception. The paper gives a brief historical review of research concerning the multiple senses of speech perception, discusses major issues, and suggests directions for future research
    Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on; 11/1996


Available from