JAIST Repository >
School of Information Science >
Conference Papers >
Conference Papers >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/16232

Title: Study on relationship between warmness of speech and valence, activation or dominance
Authors: Miyagawa, Natsumi
Akagi, Masato
Issue Date: 2020-02-20
Publisher: Research Institute of Signal Processing Japan
Magazine name: 2020 RISP International Workshop on Nonlinear Circuits, Communications and Signal Processing (NCSP 2020)
Start page: 299
End page: 302
Abstract: To communicate a robot with a warm voice like humans, it is necessary to understand the characteristics of the warm voice of humans. However, these are not explained clearly yet. In this study, we examined warm voices based on the idea of a multilayer model. We carried out three listening experiments evaluation experiment for warmness for the first layer, evaluation experiment of valence, activation, and dominance for the second layer, and evaluation experiment of semantic primitive for the third layer. Then we examined relationships between the layers of the model. As a result, it was found that the warm voice had a strong positive correlation with Valence. When dominance ratings were low or high, the warmness rating was low. If the dominance rating was close to 0, the warmness rating was high. Moreover, the warm voice had a high positive correlation with Bright and a high negative correlation with Dark, Low and Heavy.
Rights: Copyright (C) 2020 Research Institute of Signal Processing Japan. Natsumi Miyagawa and Masato Akagi, 2020 RISP International Workshop on Nonlinear Circuits, Communications and Signal Processing (NCSP 2020), 2020, pp.299-302.
URI: http://hdl.handle.net/10119/16232
Material Type: publisher
Appears in Collections:b11-1. 会議発表論文・発表資料 (Conference Papers)

Files in This Item:

File Description SizeFormat
NCSP2020_Miyagawa.pdf612KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.

 


Contact : Library Information Section, JAIST (ir-sys[at]ml.jaist.ac.jp)