Proceedings of the Annual Conference of JSAI
Online ISSN : 2758-7347
34th (2020)
Session ID : 1D5-GS-9-04
Conference information

Unsupervised Double Articulation and Meaning Acquisition of Natural Speech in a Video Game Environment
*Kotaro YAMAGUCHINatsuki OKATadahiro TANIGUCHIRyo OZAKI
Author information
CONFERENCE PROCEEDINGS FREE ACCESS

Details
Abstract

Infants need to segment their native language into phonemes and words at the same time without supervision. Taniguchi, Nagasaka, & Nakashima (2016) showed that Nonparametric Bayesian Double Articulation Analyzer could analyze latent double articulation structure, i.e., hierarchically organized latent words and phonemes, of utterance data consisting of a limited vocabulary in an unsupervised manner by assuming hierarchical Dirichlet process hidden language model (HDP-HLM). In this study, we attempted unsupervised double articulation analysis of natural speech in a video game environment and tried to give meaning to the segmented words. The result of an experiment demonstrated that the utterances were roughly correctly segmented, and the meanings of up, down, left and right were almost correctly learned.

Content from these authors
© 2020 The Japanese Society for Artificial Intelligence
Previous article Next article
feedback
Top