13th Annual Conference of the International Speech Communication Association

Portland, OR, USA
September 9-13, 2012

Detecting System-directed Utterances Using Dialogue-level Features

Kazunori Komatani (1), Akira Hirano (1), Mikio Nakano (2)

(1) Graduate School of Engineering, Nagoya University, Nagoya, Aichi, Japan
(2) Honda Research Institute Japan, Co., Ltd., Wako, Saitama, Japan

We have developed a method to determine whether a user utterance is directed at the system or not. A spoken dialogue system should not respond to audio inputs that are not directed at it (i.e., a userfs mutter), and it therefore needs to detect such inputs to avoid unsuitable responses. We classify the two cases by logistic regression based on a feature set including utterance timing, utterance length, and dialogue status. We conducted experiments using 5395 user utterances for both transcription and automatic speech recognition results. Results showed that the classification accuracy improved by 11.0 and 4.1 points, respectively. We also discuss which features are effective in the classification.

Index Terms: spoken dialogue system, system-directed utterance, utterance timing

Full Paper

Bibliographic reference.  Komatani, Kazunori / Hirano, Akira / Nakano, Mikio (2012): "Detecting system-directed utterances using dialogue-level features", In INTERSPEECH-2012, 230-233.