Etsuo Mizukami

A bootstrapping approach for SLU portability to a new language by inducting unannotated user queries

This paper proposes a bootstrapping method of constructing a new spoken language understanding (SLU) system in a target language by utilizing statistical machine translation given an SLU module in some source language. The main challenge in this work is to induct unannotated automatic speech recognition results of user queries in the source language collected through a spoken dialog system, which is under public test. In order to select candidate expressions from among erroneous translation results stemming from problems with speech recognition and machine translation, we use back-translation results to check whether the translation result maintains the semantic meaning of the original sentence. We demonstrate that the proposed scheme can effectively prefer suitable sentences for inclusion in the training data as well as help improve the SLU module for the target language.

An exploratory study for analyzing interactional processes of group discussion: the case of a focus group interview

The purposes of this study are (a) to establish a measurement for evaluating conversational impressions of group discussions, and (b) to make an exploratory investigation on their interactional processes which may affect to form those impressions. The impression rating and factor analysis undertaken first give us four factors concerning conversational impressions of “focus group interviews (FGIs)”: conversational activeness, conversational sequencing, the attitudes of participants and the relationships of participants. In relation to the factors of conversational activeness and conversational sequencing in particular, the microanalysis of four selected topical scenes from our database further shows that the behavior of the moderator and the interviewees is organized not independently but with reference to each other. The study thus emphasizes the importance of the integration of quantitative and qualitative approaches towards human interactions.

What Are They Achieving Through the Conversation? Modeling Guide–Tourist Dialogues by Extended Grounding Networks

In goal-oriented or task-oriented conversations, the participants have to share many things to achieve collaboration. The ideas of common ground, shared knowledge, and similar concepts are important to understand the process of achievement. In this study, to model guide–tourist dialogues considering such grounding process, we proposed the idea of extended grounding networks by introducing the concept of contribution topics and applied it to different data collected from dialogues between a human guide and tourists.

WFST-Based Spoken Dialogue System on Smartphones -- Its Development and Implementation for Field Use

We proposed the WFSTDM which is an expandable and adaptable dialogue management platform. The WFSTDM combines various WFSTs and enables us to develop new dialogue management WFSTs necessary for rapid prototyping of spoken dialogue systems. In this paper, we illustrate the outline of the WFSTDM and introduce the WFSTDM builder, a network-based spoken dialogue system development tool. In addition, we go into details about the spoken dialogue system AssisTra for iPhone that we developed as an example to show the implementation of the WFSTDM on smartphones. We also discuss about spoken dialogue systems on smartphones as tools for collecting field data.

NICT Kyoto Dialogue Corpus

This chapter introduces a new corpus of consulting dialogues designed for training a dialogue manager that can handle consulting dialogues through spontaneous interactions from the tagged dialogue corpus. We collected more than 150 h of consulting dialogues in the tourist guidance domain. This chapter outlines our taxonomy of dialogue act (DA) annotation that can describe two aspects of an utterance: its communicative function (speech act (SA)), and its semantic content. We provide an overview of the Kyoto tour guide dialogue corpus and a preliminary analysis using the DA tags. We also show a result of a preliminary experiment for SA tagging by Support Vector Machines (SVMs). In addition, we describe a usage of our corpus for a spoken dialogue system that we are developing.

Efficient Language Model Construction for Spoken Dialog Systems by Inducting Language Resources of Different Languages

Since the quality of the language model directly affects the performance of the spoken dialog system (SDS), we should use a statistical language model (LM) trained with a large amount of data that is matched to the task domain. When porting a SDS to another language, however, it is costly to re-collect a large amount of user utterances in the target language. We thus use the language resources in a source language by utilizing statistical machine translation. The main challenge in this work is to induct automatic speech recognition results collected using a speech-input system that differs from the target SDS both in the task and the target language. To select appropriate sentences to be included in the training data for the LM, we induct a spoken language understanding module of the dialog system in the source language. Experimental construction using over three million user utterances showed that it is vital to conduct a selection from the translation results.

Evaluating User-System Interactional Chains for Naturalness-oriented Spoken Dialogue Systems

In evaluating spoken dialogue systems, especially those aimed at actualizing a natural conversation with users, it seems necessary to assess whether the interactions between users and systems are chained appropriately. In this paper, we propose a response evaluation coding scheme for user dialogue actions and evaluate the interactional chains achieved by the users and the systems from the viewpoint of appropriateness of reactions. Two dialogue data sets of interactions between users and sightseeing guidance systems were evaluated by this scheme. The differences between the two systems and the points that should be improved were determined by analyzing the data coded by the scheme. From the result of multiple regression analysis, it was also suggested that the appropriate ratings (AR) for user actions might contribute to predicting the subjective judgment concerning naturalness of the system actions.

Sightseeing guidance systems based on WFST-based dialogue manager

We are developping a spoken dialogue system that help the users through spontaneous interactions on sightseeing guidance domain. The systems are constructed on our framework of weighted finite-state transducer (WFST) based dialogue manager. The demos are our prototype spoken dialogue systems on Kyoto tourist information assistance.