[1N1] [General Session] 2. Machine Learning

[1N1-01] Learning of Relative Spatial Concepts from Utterances based on MCMC sampling

〇Rikunari Sagara¹, Zhixiang Gu¹, Ryo Taguchi¹, Koosuke Hattori², Masahiro Hoguro², Taizo Umezaki^1,3 (1. Nagoya Institute of Technology, 2. Chubu University, 3. The University of Tokyo)

Keywords:Symbol Emergent Systems, Language Acquisition, Machine Learning

This paper presents a method for learning relative spatial concepts and phoneme sequences which represent spatial concepts and objects from utterances without knowledge of words. First, phoneme sequences recognized by a general speech recognizer are divided into words on the basis of NPYLM. Then, parameters of the relative spatial distributions are estimated from the segmented words and location information by MCMC sampling. In the experiments, the result showed that the parameters were estimated correctly by the proposed method. Moreover, phoneme sequences which represent spatial concepts and objects were learned successfully by the proposed method.