Unsupervised learning of vowel categories from infant-directed speech
TLDR
An algorithm, based on Expectation–Maximization, is presented here for learning the categories from a sequence of vowel tokens without receiving any category information with each vowel token, or knowing in advance the number of categories to learn, or having access to the entire data ensemble.Abstract:
Infants rapidly learn the sound categories of their native language, even though they do not receive explicit or focused training. Recent research suggests that this learning is due to infants' sensitivity to the distribution of speech sounds and that infant-directed speech contains the distributional information needed to form native-language vowel categories. An algorithm, based on Expectation–Maximization, is presented here for learning the categories from a sequence of vowel tokens without (i) receiving any category information with each vowel token, (ii) knowing in advance the number of categories to learn, or (iii) having access to the entire data ensemble. When exposed to vowel tokens drawn from either English or Japanese infant-directed speech, the algorithm successfully discovered the language-specific vowel categories (/i, i, e, e/ for English, /i, iː, e, eː/ for Japanese). A nonparametric version of the algorithm, closely related to neural network models based on topographic representation and competitive Hebbian learning, also was able to discover the vowel categories, albeit somewhat less reliably. These results reinforce the proposal that native-language speech categories are acquired through distributional learning and that such learning may be instantiated in a biologically plausible manner.
language acquisition
speech perception
expectation maximization
online learningread more
Citations
More filters
Journal ArticleDOI
Robust speech perception: recognize the familiar, generalize to the similar, and adapt to the novel.
TL;DR: The ideal adapter framework is formalized and can be understood as inference under uncertainty about the appropriate generative model for the current talker, thereby facilitating robust speech perception despite the lack of invariance.
Journal ArticleDOI
Category Learning in the Brain
Carol A. Seger,Earl K. Miller +1 more
TL;DR: The ability to group items and events into functional categories is a fundamental characteristic of sophisticated thought and plasticity in neural systems, including neocortical regions, the medial temporal lobe, the basal ganglia, and midbrain dopaminergic systems interact during category learning.
Journal ArticleDOI
The developmental origins of voice processing in the human brain
TL;DR: The pattern of findings suggests that temporal regions specialize in processing voices very early in development and that, already in infancy, emotions differentially modulate voice processing in the right hemisphere.
Journal ArticleDOI
What information is necessary for speech categorization? Harnessing variability in the speech signal by integrating cues computed relative to expectations
Bob McMurray,Allard Jongman +1 more
TL;DR: Even simple categorization metrics can overcome the variability in speech when sufficient information is available and compensation schemes like C-CuRE are employed.
Journal ArticleDOI
Motherese in Interaction: At the Cross-Road of Emotion and Cognition? (A Systematic Review)
Catherine Saint-Georges,Catherine Saint-Georges,Mohamed Chetouani,Raquel S. Cassel,Raquel S. Cassel,Fabio Apicella,Ammar Mahdhaoui,Filippo Muratori,Marie-Christine Laznik,David Cohen,David Cohen +10 more
TL;DR: The purpose was to provide an update of the evidence accumulated by reviewing all of the empirical or experimental studies that have been published since 1966 on IDS driving factors and impacts to suggest IDS is part of an interactive loop that may play an important role in infants’ cognitive and social development.
References
More filters
BookDOI
Density estimation for statistics and data analysis
TL;DR: The Kernel Method for Multivariate Data: Three Important Methods and Density Estimation in Action.
Journal ArticleDOI
Understanding normal and impaired word reading: computational principles in quasi-regular domains.
TL;DR: Analysis of the ability of networks to reproduce data on acquired surface dyslexia support a view of the reading system that incorporates a graded division of labor between semantic and phonological processes, and contrasts in important ways with the standard dual-route account.
Journal ArticleDOI
Cross-language speech perception: Evidence for perceptual reorganization during the first year of life
Janet F. Werker,Richard C. Tees +1 more
TL;DR: This article showed that infants can discriminate non-native speech contrasts without relevant experience, and that there is a decline in this ability during ontogeny, which is a function of specific language experience.
Journal ArticleDOI
Simplified neuron model as a principal component analyzer
TL;DR: A simple linear neuron model with constrained Hebbian-type synaptic modification is analyzed and a new class of unconstrained learning rules is derived.
Related Papers (5)
Infant sensitivity to distributional information can affect phonetic discrimination.
Cross-language speech perception: Evidence for perceptual reorganization during the first year of life
Janet F. Werker,Richard C. Tees +1 more
Cross-language analysis of phonetic units in language addressed to infants.
Patricia K. Kuhl,Jean E. Andruski,Jean E. Andruski,Inna A. Chistovich,Inna A. Chistovich,Ludmilla A. Chistovich,Ludmilla A. Chistovich,Elena V. Kozhevnikova,Elena V. Kozhevnikova,Viktoria L. Ryskina,Viktoria L. Ryskina,Elvira I. Stolyarova,Elvira I. Stolyarova,Ulla Sundberg,Ulla Sundberg,Francisco Lacerda,Francisco Lacerda +16 more