Presentation | 2021-12-22 [Poster Presentation] A Dialogue System Based on Emotion Estimation Using Deep Learning Takumi Kawakami, Shinya Fukumoto, Masayuki Kashima, Mutsumi Watanabe, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In recent years, intelligent robots have been attracting attention, and research on dialogue systems has been actively conducted as a means of communication with computers. In this paper, we propose a dialogue system based on emotion estimation. In this study, we propose a dialogue system based on emotion estimation. As an emotion estimation method, we use multiple speech features to train LSTM, a deep learning model of time series data, to estimate the user's emotions (anger, joy, sadness, surprise, calm). We also use AIML to generate rule-based response sentences that respond to the estimated emotions. We were able to obtain better accuracy in emotion estimation than in related research. In order to further improve the accuracy of emotion estimation, it is necessary to perform multimodal emotion estimation. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Dialogue systems / Emotion estimation / Deep learning / Voice |
Paper # | EA2021-59 |
Date of Issue | 2021-12-15 (EA) |
Conference Information | |
Committee | EA / US |
---|---|
Conference Date | 2021/12/22(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Sojo University |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | [Joint Meeting on Acoustics and Ultrasonics Subsociety] Engineering/Electro Acoustics, Ultrasonics, etc. |
Chair | Yoshinobu Kajikawa(Kansai Univ.) / Yoshikazu Koike(Shibaura Inst. of Tech.) |
Vice Chair | Kenichi Furuya(Oita Univ.) / Shoichi Koyama(Univ. of Tokyo) / Hikaru Miura(Nihon Univ.) / Kentaro Nakamura(Tokyo Inst. of Tech.) |
Secretary | Kenichi Furuya(NTT) / Shoichi Koyama(RitsumeikanUniv.) / Hikaru Miura(Tohoku Univ.) / Kentaro Nakamura(Chiba Univ.) |
Assistant | Yukou Wakabayashi(Tokyo Metropolitan Univ.) / Tatsuya Komatsu(LINE) / Shin Yoshizawa(Tohoku Univ.) |
Paper Information | |
Registration To | Technical Committee on Engineering Acoustics / Technical Committee on Ultrasonics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Poster Presentation] A Dialogue System Based on Emotion Estimation Using Deep Learning |
Sub Title (in English) | |
Keyword(1) | Dialogue systems |
Keyword(2) | Emotion estimation |
Keyword(3) | Deep learning |
Keyword(4) | Voice |
1st Author's Name | Takumi Kawakami |
1st Author's Affiliation | Kagoshima University(Kagoshima Univ.) |
2nd Author's Name | Shinya Fukumoto |
2nd Author's Affiliation | Kagoshima University(Kagoshima Univ.) |
3rd Author's Name | Masayuki Kashima |
3rd Author's Affiliation | Kagoshima University(Kagoshima Univ.) |
4th Author's Name | Mutsumi Watanabe |
4th Author's Affiliation | Kagoshima University(Kagoshima Univ.) |
Date | 2021-12-22 |
Paper # | EA2021-59 |
Volume (vol) | vol.121 |
Number (no) | EA-311 |
Page | pp.pp.13-17(EA), |
#Pages | 5 |
Date of Issue | 2021-12-15 (EA) |