Natural Language Dialog System Considering Speaker’s Emotion Calculated from Acoustic Features

URI http://harp.lib.hiroshima-u.ac.jp/hiroshima-cu/metadata/12354
ファイル
IWSDS2016mera.pdf ( 155.0 KB ) 公開日 :2017-12-25
タイトル
Natural Language Dialog System Considering Speaker’s Emotion Calculated from Acoustic Features
著者
氏名 TAKAHASHI Takumi
ヨミ タカハシ タクミ
別名 高橋 拓誠
氏名 MERA Kazuya
ヨミ メラ カズヤ
別名 目良 和也
氏名 TANG Ba Nhat
ヨミ
別名
氏名 KUROSAWA Yoshiaki
ヨミ クロサワ ヨシアキ
別名 黒澤 義明
氏名 TAKEZAWA Toshiyuki
ヨミ タケザワ トシユキ
別名 竹澤 寿幸
キーワード
Interactive Voice Response system (IVR)
Acoustic features
Emotion
Support Vector Machine (SVM)
Artificial Intelligence Markup Language (AIML)
抄録

With the development of Interactive Voice Response (IVR) systems, people can not only operate computer systems through task-oriented conversation but also enjoy non-task-oriented conversation with the computer. When an IVR system generates a esponse, it usually refers to just verbal information of the user’sutterance. However, when a person gloomily says “I’m fine,” people will respond not by saying “That’s wonderful” but “Really?” or “Are you OK?” because we can consider both verbal and non-verbal information such as tone of voice, facial expressions, gestures, and so on. In this paper, we propose an intelligent IVR system that considers not only verbal but also non-verbal information. To estimate a speaker’s emotion (positive, negative, or neutral), 384 acoustic features extracted from the speaker’s utterance are utilized to machine learning (SVM). Artificial Intelligence Markup Language (AIML)-based response generating rules are expanded to be able to consider the speaker’s emotion. As a result of the experiment, subjects felt that the proposed dialog system was more likable, enjoyable, and did not give machine-like reactions.

査読の有無
掲載雑誌名
Lecture Notes in Electrical Engineering
427
開始ページ
145
終了ページ
157
出版年月日
2016-12-25
出版者
Springer
ISSN
1876-1100
ISBN
978-981-10-2584-6
978-981-10-2585-3
DOI
10.1007/978-981-10-2585-3_11
本文言語
英語
資料タイプ
会議発表論文
著者版フラグ
著者版
権利情報
Copyright 2017 Springer. This is the author’s version of a work that was accepted for publication in the following source: Takumi Takahashi, Kazuya Mera, Tang Ba Nhat, Yoshiaki Kurosawa, Toshiyuki Takezawa (2017) Natural Language Dialog System Considering Speaker’s Emotion Calculated from Acoustic Features. In Kristiina Jokinen, Graham Wilcock (Eds.) Dialogues with Social Robots : Enablements, Analyses, and Evaluation, Lecture Notes in Electrical Engineering, volume 427, 145-157. The final publication is available at Springer via http://dx.doi.org/10.1007/978-981-10-2585-3_11.
関連URL
備考

This paper was previously accepted at the 7th International Workshop on Spoken Dialogue System (IWSDS2016), Saariselkä, Finland, January 13-16, 2016.
This research is supported by JSPS KAKENHI Grant Number 26330313 and the Center of Innovation Program from Japan Science and Technology Agency, JST.

区分
hiroshima-cu