Generating a Variety of Backchannels Based on Linguistic and Prosodic Features for Attentive Listening Agents
-
- YAMAGUCHI Takashi
- Kyoto University
-
- INOUE Koji
- Kyoto University
-
- YOSHINO Koichiro
- Nara Institute of Science and Technology
-
- TAKANASHI Katsuya
- Kyoto University
-
- WARD Nigel G.
- Kyoto University University of Texas at El Paso
-
- KAWAHARA Tatsuya
- Kyoto University
Bibliographic Information
- Other Title
-
- 傾聴対話システムのための言語情報と韻律情報に基づく多様な形態の相槌の生成
- ケイチョウ タイワ システム ノ タメ ノ ゲンゴ ジョウホウ ト インリツ ジョウホウ ニ モトズク タヨウ ナ ケイタイ ノ アイズチ ノ セイセイ
Search this article
Abstract
<p>There is a growing interest in conversation agents which conduct attentive listening. However, the current conversation agents always generate the same or limited form of backchannels every time, giving a monotonous impression. We have investigated generation of a variety of backchannels according to the dialogue context using the corpus of counseling dialogue. At first, we annotate all acceptable backchannel form categories considering the arbitrary nature of backchannels. Then, we conduct machine learning to predict a backchannel form from the linguistic and prosodic features of the preceding context. This model outperformed the method which always outputs the same form of backchannels and also the method which randomly generates backchannels. Finally, subjective evaluations by human listeners show that the proposed method generates backchannels more naturally giving a feeling of understanding and empathy.</p>
Journal
-
- JSAI Technical Report, SIG-SLUD
-
JSAI Technical Report, SIG-SLUD 76 (0), 09-, 2016-02-26
The Japanese Society for Artificial Intelligence
- Tweet
Details 詳細情報について
-
- CRID
- 1390007050485820544
-
- NII Article ID
- 40020797677
- 130008057692
-
- NII Book ID
- AN10432166
-
- ISSN
- 24364576
- 09185682
-
- NDL BIB ID
- 027128333
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- NDL
- CiNii Articles
-
- Abstract License Flag
- Allowed