• ETRI Journal
• /
• v.46 no.1
• /
• pp.118-126
• /
• 2024

In human conversations, listeners often utilize brief backchannels such as "uh-huh" or "yeah." Timely backchannels are crucial to understanding and increasing trust among conversational partners. In human-machine conversation systems, users can engage in natural conversations when a conversational agent generates backchannels like a human listener. We propose a method that simultaneously predicts backchannels and recognizes speech in real time. We use a streaming transformer and adopt multitask learning for concurrent backchannel prediction and speech recognition. The experimental results demonstrate the superior performance of our method compared with previous works while maintaining a similar single-task speech recognition performance. Owing to the extremely imbalanced training data distribution, the single-task backchannel prediction model fails to predict any of the backchannel categories, and the proposed multitask approach substantially enhances the backchannel prediction performance. Notably, in the streaming prediction scenario, the performance of backchannel prediction improves by up to 18.7% compared with existing methods.

• The Journal of the Korea Contents Association
• /
• v.18 no.5
• /
• pp.311-318
• /
• 2018

This paper presents the artistic background and implementation structure of a conversational artificial intelligence interactive artwork, "An Evolving GAIA". Recent artworks based on artificial intelligence technology are introduced. Development of biomimetics and artificial life technology has burred differentiation of machine and human. In this paper, artworks presenting machine-life metaphor are shown, and the distinct implementation of conversation system is emphasized in detail. The artwork recognizes and follows the movement of audience using its eyes for natural interaction. It listens questions of the audience and replies appropriate answers by text-to-speech voice, using the conversation system implemented with an Android client in the artwork and a webserver based on the question-answering dictionary. The interaction gives to the audience discussion of meaning of life in large scale and draws sympathy for the artwork itself. The paper shows the mechanical structure, the implementation of conversational system of the artwork, and reaction of the audience which can be helpful to direct and make future artificial intelligence interactive artworks.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.343-352
• /
• 2005

The existing CAI or web-based science learning system of elementary school has some disadvantages. For instance, it is composed of uniform courses designed by an instructor without considering the learner's characters, and the learner's opinions or questions raised during learning can not be delivered to the system. This structure has diminished the willingness or the motive of the learner and make an adverse effect on the learning efficiency. In this regards, Instruction-Learning System is needed to provide learning environment Pertinent to the learner's individual character and motivate the learner's active attendance and learning. This study is to design a character-based conversational Instruction-Learning System. This may induce the learner's active attendance through the communications between instructor and learner and furnish various learning materials to motivate the learners and attract their consistent interests in learning.

• Korean Journal of Cognitive Science
• /
• v.13 no.4
• /
• pp.1-8
• /
• 2002

As Internet grows, many web sites have been built, therefore much information has been registered. Because the web sites have more information, it is more difficult that the user can find the information wanted. Therefore, to get information that user wants easily, the full-text engine may be embedded to the web site. This paper is about developing the help conversational agent for a user to find the information that he wants through conversation with agent. The proposed method is based on the pattern matching of artificial intelligence, not natural language processing. If a user inputs any sentence, the help conversational agent responds to the sentence through preprocessing and pattern matching with knowledge. The knowledge is built with the XML format. With the approximate pattern matching, the agent picks up the appropriate response with some degree of similarities. At the experiment, some different sentences with the same meaning have been entered, then the agent recognized them as the same pattern, and it made a correct answer.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2049-2068
• /
• 2021

AI combined with NLP techniques has promoted the use of Virtual Assistants and have made people rely on them for many diverse uses. Conversational Agents are the most promising technique that assists computer users through their operation. An important challenge in developing Conversational Agents globally is transferring the groundbreaking expertise obtained in English to other languages. AI is making it possible to transfer this learning. There is a dire need to develop systems that understand secular languages. One such difficult language is Hindi, which is the fourth most spoken language in the world. Semantic similarity is an important part of Natural Language Processing, which involves applications such as ontology learning and information extraction, for developing conversational agents. Most of the research is concentrated on English and other European languages. This paper presents a Corpus-based word semantic similarity measure for Hindi. An experiment involving the translation of the English benchmark dataset to Hindi is performed, investigating the incorporation of the corpus, with human and machine similarity ratings. A significant correlation to the human intuition and the algorithm ratings has been calculated for analyzing the accuracy of the proposed similarity measures. The method can be adapted in various applications of word semantic similarity or module for any other language.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.620-631
• /
• 2021

The study aim to test the effect of voice agent's preceding utterance type on the user experience in the smart home contexts by conversation types. Based on two types of conversation (task-oriented vs. relationship-oriented conversations) and two types of utterance (preceding vs. response utterances), four different scenarios were designed for experimental study. A total of 62 participants were divided into two groups by utterance type, and exposed to two scenarios of the conversation types. Likeability, psychological reactance, and perceived intelligence were measured for the user experience of conversational agent. The result showed main effects of likeability in task-oriented conversations, and of psychological reactance in preceding utterances. The interaction effect demonstrated that preceding conversation improved the likeabilitty and perceived intelligence in the task-oriented conversations.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.445-455
• /
• 2022

Financial service chatbots are hindering user experience with conversational errors and machine-like responses. This study aims to examine the effect of self-disclosure and emotional expression of an anthropomorphic chatbot on user experience before conversation errors occur in financial services. In financial inquiries, scenarios were designed based on self-disclosure type (positive vs. negative) and emotional expression level(high confident vs. low confident), and online experiments were conducted. The result revealed that when anthropomorphic chatbot provided self-disclosure and emotional expression, the main effect has been shown on trust, annoyance, service recovery, and intention to continuous use. In addition, interaction effects were significant in trust and annoyance. In conclusion, this paper demonstrated that anthropomorphic chatbot's positive self-disclosure and confident emotional expression influenced trust and annoyance.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.1
• /
• pp.41-50
• /
• 2023

Recently, as the Coronavirus disease 2019 (COVID-19) prolongs along with the development of artificial intelligence technology, a non-contact society has become commonplace. Many companies are promoting digital transformation and the activation of artificial intelligence introduction to respond to this which leads to dramatic increase of demand for Chatbot. In addition, a Chatbot has reached the point of processing business transactions from responding simple inquiries. However, it is necessary to develop an API to interface with the legacy system and there are many difficulties in connecting. To solve this, it is becoming important to establish a hybrid Chatbot environment through RPA interface. Recently, the combination of RPA and Chatbot is considered an effective tool for handling many business processes. But, there are many difficulties due to the lack of interface cases and the difficulty in finding a method to development them. This study suggests a method for building a hybrid Chatbot which is an interface Chatbot(Conversational UX) and RPA(Task Automation) from the perspective of hyper-automation based on actual development cases and review of literature review is presented, so that the interface method can be understood and develop more easily. Therefore, there are implications for actively using AI Chatbot for digital transformation.

• Design Convergence Study
• /
• v.17 no.1
• /
• pp.17-28
• /
• 2018

The manipulation of the device by hand while driving is a major factor to increase the risk of accidents, and the design of in-vehicle voice conversational interface that can compensate for this is being actively researched. The purpose of this study is to investigate the effect of the use of the barge-in function of in-vehicle voice interface on user experience. Participants were asked to carry out two tasks, one for navigation and one for music play. We conducted a survey to measure the functional user 's experience after each participant' s tasks, and measured usefulness, usability, satisfaction, and emotion as user experience factors. As a result, Barge-in has been rated as the better choice for most experience factors. There was a significant effect on usability dimension in navigation task and significant effects on usability dimension and emotional dimension in music play task. So it was found that barge-in function had a positive effect on actual user's usability and emotional dimension.

• Smart Media Journal
• /
• v.12 no.6
• /
• pp.35-40
• /
• 2023

In this study paper, we examine pre-trained language models used in Multi-Goal Conversational Recommender Systems (MG-CRS), comparing and analyzing their performances of various pre-trained language models. Specifically, we investigates the impact of the sizes of language models on the performance of MG-CRS. The study targets three types of language models - of BERT, GPT2, and BART, and measures and compares their accuracy in two tasks of 'type prediction' and 'topic prediction' on the MG-CRS dataset, DuRecDial 2.0. Experimental results show that all models demonstrated excellent performance in the type prediction task, but there were notable provide significant performance differences in performance depending on among the models or based on their sizes in the topic prediction task. Based on these findings, the study provides directions for improving the performance of MG-CRS.