Artificial Intelligence Speaker Research Articles

Collaborative Learning with Artificial Intelligence Speakers

Collaborative Learning with Artificial Intelligence Speakers

Language Artificial Intelligences' Communicative Performance Quantified Through the Gricean Conversation Theory.

This study pragmatically investigates an artificial intelligence (AI) speaker (AIS)'s verbal communicative performance based on real AI-human conversation data. Specifically, this study explores Grice's conversation theory, which enables the categorization of an AIS's mistaken utterances as violations of specific conversational maxims. Twenty native Korean-speaking participants recorded at least 50 conversations with Kakao Mini AISs, provided by Daum Kakao, Inc., in Korea. Each conversation, either for information sharing or as daily dialogue, was required to contain at least two turn-taking instances. A total of 1,026 recorded dialogues were decomposed into adjacency pairs based on turn-taking. The dialogues were arranged into 3,365 adjacency pairs, and each pair was then classified as a conversational success or failure based on whether the AIS answered the user's utterance appropriately. Language users' evaluations of the AIS's mistaken expressions were also quantified via an additional acceptability rating test with 1,024 adjacency pairs. The overall results indicate that Grice's "maxim of relation" is most frequently flouted by AISs and is considered to be the least natural to language users. These findings suggest that to improve AISs' natural communication capacity, more detailed AI algorithms that generate utterances relevant to either the partner's preceding utterance or a broader conversational context should be created. Although the verbal communicative capacities of the AIS we test are substantially overtaken by those of recent large language models, such as generative pretrained transformer, the pragmatic evaluation described in the current study will remain useful for more precise linguistic quantification of current/future language AI's communicative performance/competence.

AI 스피커의 시각적 피드백 제시가 청년 및 중·장년 사용자의 지속사용의도에 미치는 영향

Artificial intelligence (AI) speakers are widely used in smart home environments, but user satisfaction is low, which can discourage user intention to continue using the device. To address this, researchers have suggested adding visual feedback to AI speakers. This study examined the effect of visual feedback on AI speaker usage intention and to verify the mediating effects of emotional experience and product attitude in younger and middle-aged adults. Videos of AI speaker disconnection situations were filmed, half of the participants were shown the videos (control group). The other half watched videos with visual feedback where context-related icons were added to the AI speakers (experimental group). For middle-aged group showed that visual feedback increased positive emotional experience, which in turn increased attitude toward the AI speaker, ultimately leading to greater intention to continue using it. However, in younger participants, emotional experience and product attitude did not mediate the relationship between visual feedback and usage intention. These findings imply that the influence of visual feedback on usage intention in user-AI speaker disconnection situations is depends on age, whereby visual feedback is especially efficacious for augmenting the usage intention of middle-aged AI speaker.

Proposal for a Sensory Integration Self-system based on an Artificial Intelligence Speaker for Children with Developmental Disabilities: Pilot Study

Proposal for a Sensory Integration Self-system based on an Artificial Intelligence Speaker for Children with Developmental Disabilities: Pilot Study

A Study on the Improvement Direction of Artificial Intelligence Speakers Applying DeLone and McLean’s Information System Success Model

Recently, many artificial intelligence speakers have been released on the market worldwide. However, the AI speaker thinks autonomously, which is a characteristic of AI. There seems to be no element to talk about. Even answering the question of this study and saying, “It’s not like an artificial intelligence speaker, it’s like a slightly advanced Internet-connected radio.” To determine the difference between these released technologies and users’ expectations in detail, we tried to analyze whether the quality and value of artificial intelligence speakers affect the positive (+) effect on use and user satisfaction. Finally, we tried to determine if it comes with user benefits. As a result of analyzing DeLone and McLean’s IS SUCCESS model, it was found that the information quality provided by the artificial intelligence speaker is used, but it is difficult to deliver satisfaction, but the service and system quality deliver user satisfaction. Moreover, it was found that perceived pleasure had no positive effect on the use or user satisfaction. It is necessary to upgrade to a level that can provide information providing quality and enjoyment that users can perceive. This means that it is a system with built-in voice recognition to answer questions connected to the Internet.

Development of the Hands-free AI Speaker System Supporting Hands-on Science Laboratory Class

The recent progress of natural language processing (NLP), speech recognition, and speech generation envisions using hands-free artificial intelligence (AI) speakers in classrooms to support student learning. In science education, the conventional hands-on laboratory education has been considered crucial in fostering students’ manipulative experimentation skills. However, touching things with gloved hands other than experimental equipment and apparatuses is strictly restricted because of the safety issue, which calls for another channel to get timely support. Therefore, we ideated that adopting hands-free AI speakers in the hands-on science laboratory classroom would support student learning. Using the rapid prototyping method, we designed and developed an AI speaker-based system that answers student queries concerning solution-making, experimental process, and waste liquid disposal, which corresponds to the initial, middle, and final phases of a laboratory class. The system was internally validated by usability tests of 9 expert panels and 18 university students, and then revised. The revised system was externally validated in an analytical chemistry experiment class for 3 sessions with 13 university students. We present the result of the prototype development, internal and external validations with quantitative and qualitative data. The AI speaker system enabled students to use the auditory learning mode in the laboratory while concentrating on the experimentation with their hands in the external validation.

Intent Classification and Slot Filling Model for In-Vehicle Services in Korean

Since understanding a user’s request has become a critical task for the artificial intelligence speakers, capturing intents and finding correct slots along with corresponding slot value is significant. Despite various studies concentrating on a real-life situation, dialogue system that is adaptive to in-vehicle services are limited. Moreover, the Korean dialogue system specialized in an vehicle domain rarely exists. We propose a dialogue system that captures proper intent and activated slots for Korean in-vehicle services in a multi-tasking manner. We implement our model with a pre-trained language model, and it includes an intent classifier, slot classifier, slot value predictor, and value-refiner. We conduct the experiments on the Korean in-vehicle services dataset and show 90.74% of joint goal accuracy. Also, we analyze the efficacy of each component of our model and inspect the prediction results with qualitative analysis.

A Study on Deriving Improvements through User Recognition Analysis of Artificial Intelligence Speakers

Recently, artificial intelligence speakers have been used a lot in homes and offices. However, users say that it is an automated speaker, not an artificial intelligence speaker. Regression analysis was performed by applying the Value-Based Acceptance Model (VAM) to see if there are any improvements to the negative perceptions of users mentioned above. As a result of the regression analysis, improvements were needed for convenience and security threats, and it did not reach the level of anthropomorphism such as with humans. In addition, it is concluded that the factors that positively affect the perceived value are usefulness and enjoyment and that they are somewhat satisfied with the burden of technical difficulties, cost, and reliability of the information. In conclusion, artificial intelligence should continuously collect various data and provide information or suggest choices and alternatives through the process of analysis, learning, and inference. However, as a result of this study, it is concluded that it is similar to an automated machine that simply finds the data among many data connected to the Internet, plays music, and connects to a site where you can shop and process it non-face-to-face. The rationale for being similar to an automated machine is that it has not reached the level of anthropomorphism.

Toward Smart Communication Components: Recent Advances in Human and AI Speaker Interaction

This study aims to investigate how humans and artificial intelligence (AI) speakers interact and to examine the interactions based on three types of communication failures: system, semantic, and effectiveness. We divided service failures using AI speaker user data provided by the top telecommunication service providers in South Korea and investigated the means to increase the continuity of product use for each type. We proved the occurrence of failure due to system error (H1) and negative results on sustainable use of the AI speaker due to not understanding the meaning (H2). It was observed that the number of users increases as the effectiveness failure rate increases. For single-person households constituted by persons in their 30s and 70s or older, the continued use of AI speakers was significant. We found that it alleviated loneliness and that human-machine interaction using AI speaker could reach a high level through a high degree of meaning transfer. We also expect AI speakers to play a positive role in single-person households, especially in cases of the elderly, which has become a tough challenge in the recent times.

AHP 분석을 통한 AI 스피커의 사용성 평가 항목에 관한 연구

Objective: The purpose of this study is to propose a usability evaluation guideline that can be primarily used in design and usability evaluation centering on artificial intelligence speakers using VUI (Voice User Interface).<BR>Background: As the artificial intelligence platform becomes the center of the Internet industry, products and services combining artificial intelligence and speech recognition technology are rapidly spreading. Among them, the artificial intelligence speaker realizes a smart home, the technology is continuously developed, Sales of artificial intelligence speakers are increasing rapidly, but the satisfaction of the artificial intelligence speaker user is still low.<BR>Method: Through the literature research and the user needs investigation, we collected and analyzed the evaluation factors of the usability of the artificial intelligence speaker to derive the factors to evaluate the usability of the artificial intelligence speaker. The derived evaluation factors were weighted through the AHP technique. The questionnaires surveys were conducted for the expert group and the user group, and the usability evaluation guideline was presented by selecting the priorities of the artificial intelligence speaker usability evaluation factors.<BR>Results: In the measurement area, analysis results showed that communication, information expression, and personalization were important attributes in order. The importance of the measurement factors showed that operability, clarity, understandability, reliability, ease of use, feedback, personality, compatibility, effectiveness, information transfer speed is important in that order.<BR>Conclusion: The usability evaluation items of artificial intelligence speaker can be considered as operability, clarity, understandability, reliability, ease of use, feedback, personality, compatibility, effectiveness, information transfer. Among these, usability evaluation should focus on factors of relatively high importance, such as operability, clarity, and understandability.<BR>Application: The results of this study are expected to be useful for the design and usability evaluation of artificial intelligence speakers.

Human likeness and attachment effect on the perceived interactivity of AI speakers

Artificial intelligence (AI) platforms with interactive assistants allow consumers to avail information and help direct their choices, thus changing the ways in which companies interact with their customers. This study aims to explore the concept of human likeness, attachment, and perceived interactivity in the context of AI speakers and examine the effects of human likeness and attachment on their perceived interactivity with AI speakers. A survey was used to investigate whether consumers’ perception of human likeness, attachment and their perceived interactivity with AI speakers eventually influence their purchase intention, and how attachment and the perceived interactivity mediate the effects of human likeness and purchase intention. The participants were 311 AI speaker users from South Korea. The results demonstrate how the direct effect and mediating effect of human likeness, attachment, and perceived interactivity on AI speakers boost purchase intention, thus providing meaningful implications for AI-driven marketing academics and practitioners.

AI-powered recommendations: the roles of perceived similarity and psychological distance on persuasion

Artificial intelligence (AI) plays various roles in our daily lives, such as personal assistant, salesperson, and virtual counselors; thus, it stands out in various fields as a recommendation agent. This study explored the effects of perceived similarity and psychological distance on the persuasion of AI recommendation agents through two experiments. Results of Experiment 1 elucidated that individuals feel more psychologically distant when they interact with AI recommendation agents than with human agents as a result of a different level of perceived similarity. Furthermore, psychological distance plays a mediating role in determining the effectiveness of desirability- vs. feasibility-focused messages in health-related issues. In Experiment 2, we manipulated the AI speaker's level of perceived similarity via anthropomorphism and found that the AI's recommendation with secondary (vs. primary) features is more effective when AI is humanized, and the reverse was found in non-humanized AI conditions. Both theoretical and managerial implications are provided.

인공지능 스피커의 디스플레이 탑재가 사용자 경험에 미치는 영향

인공지능 스피커의 디스플레이 탑재가 사용자 경험에 미치는 영향

Quality evaluation model of artificial intelligence service for startups

PurposeThis paper aims to develop a quality evaluation model for artificial intelligence (AI)-based products/services that is applicable to startups utilizing AI technology. Although AI-based service has risen dramatically and replaced many service offerings, in reality, startups are rarely to develop and evaluate AI services. The features of AI service are fundamentally different from the properties of existing services and have a great influence on the customer's service selection.Design/methodology/approachThis paper reviews startups' development process, existing quality evaluation models and characteristics of services utilizing AI technology, and develops a quality evaluation model for AI-based services. A detailed analysis of a survey (application of the model) on customer satisfaction for AI speakers is provided.FindingsThis paper provides seven key features and 24 evaluation items for evaluating AI-based services.Originality/valueThis paper contributes to the growing need for methodologies that reflect the new era of AI-based products/services in quality evaluation research.

A Study on Consumer Strategy of Artificial Intelligence Service Using Importance-Satisfaction Analysis

Purpose: The purpose of this paper is to investigate satisfaction and importance in order to appropriately evaluate AI-based service quality evaluation. It is to conduct an importance and satisfaction analysis for users of artificial intelligence speakers and derive a service strategy. Design/methodology/approach: This study conducted a survey of consumers using artificial intelligence speakers in Republic of Korea. The survey was conducted online and the total number of samples was 200. Findings: In this study, meaningful results were derived by conducting an overall importance-satisfaction analysis targeting artificial intelligence speaker users. Based on this, a maintenance strategy, improvement strategy, and reinforcement strategy were established for each quality item of artificial intelligence service. By analyzing gender characteristics, service strategies suitable for men and women were derived. Research limitations/implications: It is necessary to derive a broad service strategy by conducting additional analysis based on different criteria such as by brand. In the future, research is needed to expand to various services based on artificial intelligence in addition to artificial intelligence speakers. Originality/value: These research results will help to establish appropriate strategies for various services based on artificial intelligence.

Analysis of User Experience and Usage Behavior of Consumers Using Artificial Intelligence(AI) Devices

Artificial intelligence (AI) devices are rapidly emerging as a core platform of next-generation information and communication technology (ICT), this study investigated consumer usage behavior and user experience through AI devices that are widely applied to consumers' daily lives. To this end, data was collected from 600 consumers with experience in using AI devices were derived to recognize the attributes and behavior of AI devices. The analysis results are as follows. First, music listening was the most used among various attributes and it was found that simple functions such as providing weather information were usefully recognized. Second, the main devices used by AI device users were identified as AI speakers, smartphone, PC and laptops. Third, associative images of AI devices appeared in the order of fun, useful, novel, smart, innovative, and friendly. Therefore, practical implications are suggested to contribute to provision of user services using AI devices in the future by analyzing usage behaviors that reflect the characteristics of AI devices.

A Redundant Sensing Elimination Technique for Improving Energy Efficiency of IoT Sensor Networks

Various sensor nodes used in IoT network systems composed of electronic devices perform various control tasks using the results measured by the sensors. At this point, if these sensing tasks are smartly assigned and performed, the energy consuming efficiency of the entire networked system can be improved. That is, since each sensor node constituting the IoT network system performs redundant sensing and control devices, it is possible to improve the network’s energy efficiency by appropriately assigning and removing such redundant sensing tasks. In particular, the indoor illuminance sensor is installed in various devices such as automatic light switch systems, smart phones, and tablets. It is desirable that only one sensor node among them participates in the actual sensing task. In a smart home, redundant sensors and controllers are used repeatedly in refrigerators, heaters, TVs, artificial intelligence speakers, ventilation facilities, and smartphones. These redundant sensing and control signal generation lowers energy consumption’s efficiency. This study proposes a technique that can properly eliminate unnecessary and redundant sensing tasks that occur when such on the flown IoT network system is built. As a result, it has been shown that the efficiency of energy consumption can be improved in the entire IoT sensor network system.

네트워크 외부성에 따른 소비자 유형화 연구: 인공지능 스피커를 중심으로

네트워크 외부성에 따른 소비자 유형화 연구: 인공지능 스피커를 중심으로

The Display Week 2021 Symposium Preview

The Display Week 2021 Symposium Preview

A Study on the Use of Artificial Intelligence Speakers for the People with Physical disability using Technology Acceptance Model

A Study on the Use of Artificial Intelligence Speakers for the People with Physical disability using Technology Acceptance Model