Models of Phonetic Recognition III: The Role of Analysis by Synthesis in Phonetic Recognition

Authors

  • Victor W. Zue

Abstract

This paper proposes a recognition model that attempts to deal with variabilities found in the acoustic signal. The input speech signal is first transformed into a representation that takes into account known properties of the human auditory system. From various stages of this transformation, acoustic parameters are extracted and used to classify the utterance into broad phonetic categories. The outcome of this analysis is used for lexical access. The constraints imposed by the language on possible sound patterns should significantly reduce the number of word candidates. Finally, detailed acoustic cues will be utilised to select the correct word from the small set of candidate words.

Additional Files

Published

2022-12-03

How to Cite

1.
Zue VW. Models of Phonetic Recognition III: The Role of Analysis by Synthesis in Phonetic Recognition. Canadian Acoustics [Internet]. 2022 Dec. 3 [cited 2024 Nov. 24];14(3 bis):69-70. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/3526

Issue

Section

Proceedings of the Acoustics Week in Canada