lantillas para corte de probetas - adendorf · go también de acero para poder sostener firmemente...

Cue Integration With Categories: Weighting Acoustic Cues in Speech Using Unsupervised Learning and Distributional Statistics Joseph C. Toscano, Bob McMurray Department of Psychology and Delta Center, University of Iowa Received 24 November 2008; received in revised form 15 May 2009; accepted 3 August 2009 Abstract During speech perception, listeners make judgments about the phonological category of sounds by taking advantage of multiple acoustic cues for each phonological contrast. Perceptual experiments have shown that listeners weight these cues differently. How do listeners weight and combine acoustic cues to arrive at an overall estimate of the category for a speech sound? Here, we present several simulations using a mixture of Gaussians models that learn cue weights and combine cues on the basis of their distributional statistics. We show that a cue-weighting metric in which cues receive weight as a function of their reliability at distinguishing phonological categories provides a good ﬁt to the perceptual data obtained from human listeners, but only when these weights emerge through the dynamics of learning. These results suggest that cue weights can be readily extracted from the speech signal through unsupervised learning processes. Keywords: Speech perception; Speech development; Cue weighting; Reliability; Categorization; Statistical learning; Unsupervised learning; Mixture of Gaussians 1. Introduction In every domain of perception, multiple sources of information must be combined. A classic example is depth perception, where the distance of an object from an observer is indicated by a number of cues, including stereopsis, perspective, binocular disparity, shad- ing, motion, and many others (see Kaufman, 1974, for an extensive list). We use the term cue here to refer to any source of information that allows the perceiver to distinguish Correspondence should be sent to Joseph Toscano, Department of Psychology, E11 SSH, University of Iowa, Iowa City, IA 52242. E-mail: [email protected] Cognitive Science 34 (2010) 434–464 Copyright Ó 2009 Cognitive Science Society, Inc. All rights reserved. ISSN: 0364-0213 print / 1551-6709 online DOI: 10.1111/j.1551-6709.2009.01077.x

lantillas para corte de probetas - adendorf · go también de acero para poder sostener firmemente...

Documents