Adverse conditions improve distinguishability of auditory, motor and perceptuo-motor theories of speech perception: an exploratory Bayesian modeling study

Moulin-Frier, Clément and Laurent, Raphaël and Bessière, Pierre and Schwartz, Jean-Luc and Diard, Julien

Link to the article

Abstract

In this paper, we put forward a computational framework for the comparison between motor, auditory and perceptuo-motor theories of speech communi-cation. We first recall the basic arguments of these three sets of theories, ei-ther applied to speech perception or to speech production. Then we expose a unifying Bayesian model able to express each theory in a probabilistic way. Focusing on speech perception, we demonstrate that under two hypotheses, regarding communication noise and inter-speaker variability, providing per-fect conditions for speech communication, motor and auditory theories are indistinguishable. We then degrade successively each hypothesis to study the distinguishability of the different theories in "adverse" conditions. We first present simulations on a simplified implementation of the model with mono-dimensional sensory and motor variables, and secondly we consider a simula-tion of the human vocal tract providing more realistic auditory and articula-tory variables. Simulation results allow us to emphasize the respective roles of motor and auditory knowledge in various conditions of speech perception in adverse conditions, and to suggest some guidelines for future studies aiming at assessing the role of motor knowledge in speech perception.

Bibtex

@article{moulinfrier2012adverse,
  title = {{Adverse conditions improve distinguishability of auditory, motor and perceptuo-motor theories of speech perception: an exploratory Bayesian modeling study}},
  author = {Moulin-Frier, Cl{\'e}ment and Laurent, Rapha{\"e}l and Bessi{\`e}re, Pierre and Schwartz, Jean-Luc and Diard, Julien},
  journal = {Language and Cognitive Processes},
  year = {2012},
  number = {7-8, Special Issue: Speech Recognition in Adverse Conditions},
  pages = {1240-1263},
  volume = {27},
  affiliation = {Grenoble Images Parole Signal Automatique - GIPSA-lab , Laboratoire d'Informatique de Grenoble - LIG , Laboratoire de psychologie et neurocognition - LPNC},
  doi = {10.1080/01690965.2011.645313 },
  hal_id = {hal-01059179},
  keywords = {Auditory, Motor and Perceptuo-motor theories of speech communication, Bayesian modeling, speech perception in adverse conditions, model distin-guishability.},
  language = {English}
}