Presentation | 2021-06-18 [Invited Talk] Spoken Dialogue System for Android ERICA Koji Inoue, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | One of the major goals for artificial intelligence research is to realize machines (or robots) that can talk like human beings. Applying deep learning to spoken language processing has advanced its technology dramatically, with spoken dialogue systems being put to practical use in various situations and becoming more commonplace. However, since current dialogue systems are still rather mechanical and simple such as question and answering, it is hard to say that the systems have achieved "human-level deep conversation". In this talk, I introduce a research project involving autonomous android ERICA and its spoken dialogue system. The ultimate goal of this research project is to realize spoken dialogue comparable to that done by human beings. Firstly, I explain basic prediction models of backchannels and turn-taking to improve human-level dialogue. Then I describe the implementation and dialogue experiments for three specific spoken dialogue tasks: attentive listening, job interview, and laboratory guide. For attentive listening, a multimodal Turing test was conducted to compare the system with human listeners (Wizard-of-Oz). The results showed that the system achieved the same level of performance as humans in some basic aspects of attentive listening but currently is not at the level of deeper conversation. Finally, reviewing the challenges found in this project, future directions for research on spoken dialogue systems are proposed. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Spoken dialogue system / Android / Turing test |
Paper # | SP2021-7 |
Date of Issue | 2021-06-11 (SP) |
Conference Information | |
Committee | SP / IPSJ-SLP / IPSJ-MUS |
---|---|
Conference Date | 2021/6/18(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | OTOGAKU Symposium 2021 |
Chair | Hisashi Kawai(NICT) / 北岡 教英(豊橋技科大) / 竹川 佳成(はこだて未来大) |
Vice Chair | |
Secretary | (Univ. of Tokyo) / (Waseda Univ.) / (京大) |
Assistant | Yusuke Ijima(NTT) |
Paper Information | |
Registration To | Technical Committee on Speech / Special Interest Group on Spoken Language Processing / Special Interest Group on Music and Computer |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Invited Talk] Spoken Dialogue System for Android ERICA |
Sub Title (in English) | A Multimodal Turing Test Challenge |
Keyword(1) | Spoken dialogue system |
Keyword(2) | Android |
Keyword(3) | Turing test |
1st Author's Name | Koji Inoue |
1st Author's Affiliation | Kyoto University(Kyoto Univ.) |
Date | 2021-06-18 |
Paper # | SP2021-7 |
Volume (vol) | vol.121 |
Number (no) | SP-66 |
Page | pp.pp.37-37(SP), |
#Pages | 1 |
Date of Issue | 2021-06-11 (SP) |