Sound objects: a bio-inspired representation, hierarchical sparse to very large dimensions used in recognition

Authors

  • Simon Brodeur Groupe de recherche en Neuroscience Computationelle et Traitement Intelligent des Signaux (NECOTIS) Département génie électrique et génie informatique, Université de Sherbrooke, Sherbrooke QC Canada J1K 2R1
  • Jean Rouat Groupe de recherche en Neuroscience Computationelle et Traitement Intelligent des Signaux (NECOTIS) Département génie électrique et génie informatique, Université de Sherbrooke, Sherbrooke QC Canada J1K 2R1

Abstract

The emphasis is put on the hierarchical structure, independence and sparseness aspects of auditory signalrepresentations in high-dimensional spaces, so as to define the components of auditory objects. The conceptof an auditory object and its neural representation is introduced. An illustrative application then follows,consisting in the analysis of various auditory signals : speech, music and natural outdoor environments. Anew automatic speech recognition (ASR) system is then proposed and compared to a conventional statisticalsystem. The proposed system clearly shows that an object-based analysis introduces a great flexibility androbustness for the task of speech recognition. The integration of knowledge from neuroscience and acousticsignal processing brings new ways of thinking to the field of classification of acoustic signals.

Published

2013-06-01

How to Cite

1.
Brodeur S, Rouat J. Sound objects: a bio-inspired representation, hierarchical sparse to very large dimensions used in recognition. Canadian Acoustics [Internet]. 2013 Jun. 1 [cited 2021 Sep. 21];41(2):37-52. Available from: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/2617

Issue

Section

Article - Signal Processing / Numerical Methods