• Publications
  • Influence
Designing Precise and Robust Dialogue Response Evaluators
This work proposes to build a reference-free evaluator and exploit the power of semi-supervised training and pretrained (masked) language models and achieves a strong correlation with human judgement and generalizes robustly to diverse responses and corpora. Expand
Improved End-to-End Speech Emotion Recognition Using Self Attention Mechanism and Multitask Learning
A speech emotion recognition (SER) method using end-to-end (E2E) multitask learning with self attention to deal with several issues is proposed, which outperforms the state-of-the-art methods and improves the overall accuracy. Expand
A Conversational Dialogue Manager for the Humanoid Robot ERICA
A dialogue system for a conversational robot, Erica, that integrates question-answering with a statement response component which generates dialogue by asking about focused words detected in the user’s utterance, and a proactive initiator which generates Dialogue based on events detected by Erica. Expand
Effective Incorporation of Speaker Information in Utterance Encoding in Dialog
A relative speaker modeling method is proposed to address the problem where speaker annotations are inconsistent among different dialogs and results show that the proposed method yields superior and more consistent performances. Expand
Talking with ERICA, an autonomous android
We demonstrate dialogues with an autonomous android ERICA, who has an appearance like a human being. Currently, ERICA plays two social roles: a laboratory guide and a counselor. It is designed toExpand
Joint dialog act segmentation and recognition in human conversations using attention to dialog context
The proposed encoder–decoder model featuring joint coding and incorporate contextual information by means of an attentional mechanism outperforms other models in segmentation, and the application of attentions significantly reduces recognition error rates. Expand
Joint Learning of Dialog Act Segmentation and Recognition in Spoken Dialog Using Neural Networks
A unified architecture for these two tasks is investigated, which aims to improve the model’s performance on both of the tasks, and shows that the joint training system outperforms the simple cascading system and the joint coding system on both dialog act segmentation and recognition tasks. Expand
Interactional and pragmatics-related prosodic patterns in mandarin dialog
The roles of prosody vary from language to language. In European languages prosody is largely involved in pragmatics, but this may be less true for other languages, especially tone languages. As aExpand
Multimodal interaction with the autonomous Android ERICA
In this demonstration the user can converse with ERICA on a number of topics and the dialog management system and the eye gaze behavior of ERICA used for indicating attention and turn taking are demonstrated. Expand
A Unified Neural Architecture for Joint Dialog Act Segmentation and Recognition in Spoken Dialog System
A unified architecture based on neural networks is proposed, which consists of a sequence tagger for segmentation and a classifier for recognition of dialog act units and shows that the jointly-trained models outperform independently- trained models, single-step models, and other reported results in DA segmentation, recognition, and joint tasks. Expand