Home Assistance Research Articles

A Deep Learning Approach for Speech Emotion Recognition Optimization Using Meta-Learning

Speech emotion recognition (SER) is widely applicable today, benefiting areas such as entertainment, robotics, and healthcare. This emotional understanding enhances user-machine interaction, making systems more responsive and providing more natural experiences. In robotics, SER is useful in home assistance devices, eldercare, and special education, facilitating effective communication. Additionally, in healthcare settings, it can monitor patients’ emotional well-being. However, achieving high levels of accuracy is challenging and complicated by the need to select the best combination of machine learning algorithms, hyperparameters, datasets, data augmentation, and feature extraction methods. Therefore, this study aims to develop a deep learning approach for optimal SER configurations. It delves into the domains of optimizer settings, learning rates, data augmentation techniques, feature extraction methods, and neural architectures for the RAVDESS, TESS, SAVEE, and R+T+S (RAVDESS+TESS+SAVEE) datasets. After finding the best SER configurations, meta-learning is carried out, transferring the best configurations to two additional datasets, CREMA-D and R+T+S+C (RAVDESS+TESS+SAVEE+CREMA-D). The developed approach proved effective in finding the best configurations, achieving an accuracy of 97.01% for RAVDESS, 100% for TESS, 90.62% for SAVEE, and 97.37% for R+T+S. Furthermore, using meta-learning, the CREMA-D and R+T+S+C datasets achieved accuracies of 83.28% and 90.94%, respectively.

Perceived creepiness in response to smart home assistants: A multi-method study

Smart home assistants (SHAs) have gained a foothold in many households. Although SHAs have many beneficial capabilities, they also have characteristics that are colloquially described as creepy – a fact that may deter potential users from adopting and utilizing them. Previous research has examined SHAs neither from the perspective of resistance nor the perspective of creepiness. The present research addresses this gap and adopts a multi-method research design with four sequential studies. Study 1 serves as a pre-study and provides initial exploratory insights into the concept of creepiness in the context of SHAs. Study 2 focuses on developing a measurement instrument to assess perceived creepiness. Study 3 uses an online experiment to test the nomological validity of the construct of creepiness in a larger conceptual model. Study 4 further elucidates the underlying behavioral dynamics using focus group analysis. The findings contribute to the literature on the dark side of smart technology by analyzing the triggers and mechanisms underlying perceived creepiness as a novel inhibitor to SHAs. In addition, this study provides actionable design recommendations that allow practitioners to mitigate end users’ potential perceptions of creepiness associated with SHAs and similar smart technologies.

Use of Artificial Intelligence Techniques to Assist Individuals with Physical Disabilities.

Assistive technologies (AT) enable people with disabilities to perform activities of daily living more independently, have greater access to community and healthcare services, and be more productive performing educational and/or employment tasks. Integrating artificial intelligence (AI) with various agents, including electronics, robotics, and software, has revolutionized AT, resulting in groundbreaking technologies such as mind-controlled exoskeletons, bionic limbs, intelligent wheelchairs, and smart home assistants. This article provides a review of various AI techniques that have helped those with physical disabilities, including brain-computer interfaces, computer vision, natural language processing, and human-computer interaction. The current challenges and future directions for AI-powered advanced technologies are also addressed.

An Interaction Framework for Designing Systems for Virtual Home Assistants and People with Dysarthria

AbstractPeople with speech impairments usually use assistive technology devices to assist them with communication and daily tasks. These devices can be controlled using different modalities, such as touch, eye gaze, gestures, and others. This article proposes a standardized methodology for designing non-verbal voice cue interactive systems that enable people with dysarthria to vocally interact with virtual home assistants (VHAs). We adopted a qualitative data-gathering approach to gain insights into users’ experiences and requirements and to determine crucial design elements for designing interactive voice assistants for people with dysarthria. Nineteen participants with varying levels of dysarthria took part in the study to create the framework. A system was then built using the proposed framework, and an additional test was performed with a further seven participants to validate the created system, thus inferring the validity of the framework. Our work empirically demonstrates how an informed, structured design of a fast, direct (verbal rather than forcing users to change modalities or using an intermediate device) method of communication improves the usability of VHAs for people with dysarthria while simultaneously allowing for a more authentic experience. The current data also highlighted that using non-verbal voice cues would be a convenient option. By providing a reproducible framework for developing non-verbal interactive systems for VHAs, we can increase the accessibility of said devices, their usability, and their user experience.

Understanding the needs of older adults learning to use digital home assistants: A demonstration study

Understanding the needs of older adults learning to use digital home assistants: A demonstration study

The Role of Recurrent Convolutional Neural Network in IoT for Building a Security Artificial Intelligence and Home Assistance System

The Role of Recurrent Convolutional Neural Network in IoT for Building a Security Artificial Intelligence and Home Assistance System

Pré-implémentation d’indicateurs de fragilité et de complexité dans les soins à domicile : une analyse à l’aide de l’Implementation Research Logic Model

Introduction : Des indicateurs cliniques de fragilité et de complexité des situations de soins à domicile sont dérivés de l’interRAI HC, mais leur implémentation dans la pratique demande une analyse précise et préalable du contexte de soins dans lequel ils pourraient être utilisés. L’Implementation Research Logic Model (IRLM) est un cadre de référence mobilisable pour l’ensemble du processus d’implémentation notamment pour la pré-implémentation, une étape indispensable, bien que souvent négligée. Objectif : Cet article vise à présenter une démarche d’application du modèle IRLM qui précise les éléments clés (déterminants, stratégies, mécanismes et retombées) de l’étape de pré-implémentation. Démarche : L’implémentation prend place à l’institution genevoise de maintien à domicile (imad) en Suisse. Les données recueillies (ex. caractéristiques de l’institution, questionnaire auprès des infirmières) sont classées dans l’IRLM par un groupe de travail. Évaluation : L’analyse des éléments de pré-implémentation pointe les déterminants contextuels pouvant favoriser l’implémentation (ex. disponibilité des indicateurs cliniques et protocoles de prise en soins) ou la freiner (ex. multiplicité des informations). Par ailleurs, les stratégies mises en place et les retombées recueillies à ce stade (ex. intégration des indicateurs dans le dossier du patient, peu de résistance au changement) sont décrites. Discussion et conclusion : L’anticipation et l’analyse des conditions dans lesquelles prend place l’implémentation dès la phase de pré-implémentation permettent de recueillir des informations essentielles sur des recommandations d’adaptation des actions aux contextes spécifiques et aux réalités concrètes. La mise en place de stratégies étape par étape devrait contribuer à relever le défi du transfert de nouvelles connaissances à la pratique clinique contextualisée.

How does anthropomorphism improve human-AI interaction satisfaction: a dual-path model

With the development of artificial intelligence technology, the application of anthropomorphism has been established as a good design principle. However, the effectiveness of anthropomorphism in smart home assistants is still unclear, as is the mechanism influencing interaction satisfaction. Therefore, we have categorized the anthropomorphic cues of smart home assistants into four dimensions: visual, identity, emotional, and auditory, and constructed a dual-path influence mechanism model of anthropomorphism on interaction satisfaction. Furthermore, we also consider the different influences of different usage contexts. Through two experimental studies, the results show that, in the intimacy influence path, the emotional cue and auditory cue can directly improve interaction satisfaction, and they can further improve interaction satisfaction by affecting intimacy. However, the visual cue has no significant direct effect on interaction satisfaction. The visual cue mainly improves interaction satisfaction through the intermediate effect of intimacy. In the privacy invasion influence path, the emotional cue and auditory cue also further affect interaction satisfaction by affecting privacy invasion. However, the impact of the visual cue and identity cue on privacy invasion was not valid. The study is expected to provide practical guidance for product designers to realize the long-term feasibility and sustainability of products.

Technology-based Project-based Learning to Teach English in University Level: Need Analysis

The goal of this study is to create a conceptual framework for implementing project-based learning in private institutions to teach language skills. This study made use of the LPKS Aura Sukma Insani-Tourism and the Home Assistance Program in the English study program. An explanatory mixed methods design method was used in this study, with quantitative research and qualitative analysis conducted to support the analysis. The needs analysis model is a hybrid of CIPP and OEM evaluation concepts. According to the findings of this study, this institution in this context requires a conceptual framework for the use of technology-based Project-based learning in teaching language skills, so that trainers are pedagogically prepared. The input results indicate that this institution requires trainers and students to be prepared to implement technology-based Project-based learning in the classroom. The process results indicate that technology-based project-based learning must be implemented with adequate technological support for students. The product results demonstrate that this technology-based project-based learning must be capable of producing results in the form of products that will later be used as evaluation material in the implementation of additional technology-based project-based learning. In addition to the findings of the analysis, it was discovered that the planning of teaching materials must be in accordance with student needs as well as 21st century skills.
 
 Keywords: language skill, project-based learning, technology, 21st century learning

Breastfeeding practices based on the gestational age and weight at birth in the first six months of life in a population-based cohort of infants from North India.

Short and long term benefits of early Initiation of breastfeeding (EIBF) and exclusive breastfeeding (EBF) in the first six months of life are well established and recommended globally. However, reliable estimates of breastfeeding practices and impact of breastfeeding counselling interventions according to gestational age and weight at birth are not available in low and middle income countries. To assess the impact of breastfeeding counselling on EIBF and EBF during the first 6 months of life according to gestational age and weight at birth. We analysed the data collected from the Women and Infants Integrated Interventions for Growth Study (WINGS), an individually randomized factorial design trial. Mothers were counselled on EIBF during third trimester of pregnancy. They were supported throughout the first 6 months to continue EBF by early problem identification, frequent home visits and assistance in expressing breastmilk when direct breastfeeding was not possible. Breastfeeding practices were ascertained through 24 h recalls at infant ages 1, 3 and 5 months for both the intervention and control groups by an independent outcome ascertainment team. The World Health Organization (WHO) definitions were used for classification of infant breastfeeding practices. Generalized linear models of the Poisson family with a log-link function were used to estimate the effect of interventions on breastfeeding practices. The relative measures of effect on breastfeeding practices were estimated in term appropriate for gestational age (T-AGA), term small for gestational age (T-SGA), preterm AGA (PT-AGA), preterm SGA (PT-SGA) infants. Amongst all infants irrespective of gestational age and weight at birth, EIBF was (51.7%) higher amongst the intervention group (IRR 1.38, 95% CI 1.28-1.48) compared with the control group. The proportion of exclusively breastfed infants at ages 1 month (IRR 1.37, 95% CI 1.28-1.48), 3 months (IRR 2.13, 95% CI 1.30-1.44) and 5 months (IRR 2.78, 95% CI 2.58-3.00) were higher in intervention group than control group. We identified significant interaction (p value for interaction <0.05) between intervention and infant size and gestation at birth on exclusive breastfeeding at 3 and 5 months of age. Subgroup analysis showed that the impact of the intervention was greater on exclusive breastfeeding in PT- SGA infants at 3 months (IRR 3.30, 95% CI 2.20-4.96) and 5 months of age (IRR 5.26, 95% CI 2.98-9.28). This is one of the first studies wherein impact of breastfeeding counselling interventions in the first 6 months of life was assessed according to infant size and gestation at birth wherein gestational age was reliably estimated. The impact of this intervention was higher in preterm and SGA babies compared to other infants. This finding is important as preterm and SGA infants have a higher burden of mortality and morbidity during early infancy. Intensive breastfeeding counselling to these vulnerable infants is likely to improve overall breastfeeding rates and reduce the adverse outcomes.Clinical Trial Registration: [http://ctri.nic.in/Clinicaltrials/pmaindet2.php?trialid=19339%26EncHid=%26userName=societyforappliedstudies], identifier [#CTRI/2017/06/008908].

Evaluation of IoT-Based Smart Home Assistance for Elderly People Using Robot

In the development of Internet-of-things (IoT)-based technology, there is a pre-programmed robot called Cyborg which is used for assisting elderly people. It moves around the home and observes the surrounding conditions. The Cyborg is developed and used in the smart home system. The features of a smart home system with IoT technology include temperature control, lighting control, surveillance, security, smart electricity, and water sensors. Nowadays, elderly people may not be able to maintain their homes appropriately and may feel uncomfortable performing day-to-day activities. Therefore, Cyborg can be used to carry out the activities of elderly people. Such activities include switching off unnecessary lights, watering plants, gas control, monitoring intruders or unknown persons, alerting elderly people in emergency situations, etc. These activities are controlled by web-based platforms as well as smartphone applications. The issues with the existing algorithms include that they are inefficient, require a long time for implementation, and have high storage space requirements. This paper proposes the k-nearest neighbors (KNN) with an artificial bee colony (ABC) algorithm (KNN-ABC). In this proposed work, KNN-ABC is used with wireless sensor devices for perceiving the surroundings of the smart home. This work implements the automatic control of electronic appliances, alert signal processors, intruder detection, and performs day-to-day activities automatically in an efficient way. GNB for intruder detection in the smart home environment system using the Cyborg human intervention robot achieved an accuracy rate of 88.12%, the Artificial Bee Colony algorithm (ABC) achieved 90.12% accuracy on the task of power saving in smart home electronic appliances, the KNN technique achieved 91.45% accuracy on the task of providing a safer pace to the elderly in the smart home environment system, and our proposed KNN-ABC system achieved 93.72%.

Using Virtual Home Assistants to Address Vulnerable Adults' Complex Care Needs: An Exploratory Feasibility Study.

Harnessing technology has been proposed as one strategy to meet the social and health needs of older adults who prefer to age in place, but solutions remain elusive. In the current study, we evaluated the feasibility of using voice-activated virtual home assistants (VHAs; i.e., Amazon Echo "Alexa" devices) with older adults participating in the Program for All-Inclusive Care of the Elderly (PACE) over 4 months. Study methods included process evaluations, tracking participants' VHA use, and qualitative feedback from PACE participants and staff. The most common VHA activities were voice-activated smart lighting and asking for information. Participants infrequently used VHA activities that could promote physical or cognitive function (e.g., chair yoga, word recall game). Participants enjoyed using the VHAs, and PACE staff were enthusiastic about the potential for VHAs to facilitate aging in place and provided recommendations to increase participants' use of functional health activities. [Journal of Gerontological Nursing, 49(6), 33-40.].

An approach for Correcting the Word-level Mispronunciations for non-native English-speaking Indian Children

Speech Recognition and its potential applications in terms of “talking devices” have become indispensable in today’s world. Technological advances like mobiles, smart home assistants or tablets extensively use the techniques of automatic speech recognition that works good for adults but cannot always follow and understand children’s speech. The primary goal of this paper is to bridge the gap of communication between voice assistants and Indian children speaking English as secondary language. The issue of lack of children’s speech corpora with English as non-native language, is addressed by creating a dataset of children in the age group of 5-15 years, speaking Hindi or Marathi as their mother tongue and English as their second language. The analysis and implementation of the proposed work shows the accuracy of approximately 96% and potential for further scope by increasing the size of dataset in lower age group. The key contributions of our work are (i) creating speech dataset of Indian children whose mother-tongue is Hindi or Marathi, (ii) employing and evaluating hybrid Convolutional Neural Network (CNN) as an age classifier, (iii) language modeling to customize children vocabulary, (iv) checking accuracy and performance of the system.

From Screens to Projector, Wall, and TVs: Conceptualizing Livestreams as Design Material for Direct and Indirect Viewership Experiences

Livestreams are now ubiquitous-moving beyond web and mobile-based platforms into products such as fitness machines and smart home assistants. As such, viewership practices have also become varied and diverse: livestreams are used for interactive entertainment, social companionship, and multi-perspective spectatorship for music festivals and sporting events. At the premier in-person gaming livestream event, Games Done Quick (GDQ), attendees engage in a variety of different viewership practices across different event spaces where the livestream is projected on large stage screens, onto walls as peripheral displays, and routed to televisions in hotel venue rooms. This study takes GDQ as an information-rich fieldsite to examine how livestream viewership shifts, changes, and interleaves across physical spaces to provide an expanded livestream experience. From 11 semi-structured interviews with GDQ attendees and participant observations at three GDQ events, we found diverse practices of both direct and indirect viewership: the livestream is viewed in collocation with an audience and stage, but is also used to establish viewership continuity across disjoint spaces, tangentially to shop for interesting moments, as a timekeeping device, and even as a means of auditory obfuscation. To unpack these forms of viewership, we conceptualize livestreams as design material and identify its two key qualities of focality and metacontent. We illustrate how examining livestreams' focality and metacontent qualities allow us to understand the relationship between different livestream configurations and different viewership behaviors, as well as chart out new under-explored design directions for livestreaming that blend physical and digital spaces.

Foni Drone

Voice control technology has revolutionized the way we interact with devices, and drone technology is no exception,through this system by the voice commands with any physical machine allows the human beings to acquiring new devices. This technology is useful in future to do certain task without physical human interaction. Technology to help them in daily life, control things via voice like a home assistant in a comfortable manner. This system will be useful to rescue, surveillance and many other areas which generally use drones as their operations. Key Words: ESP-32, gyroscope

Distance Learning Home Experience and the Food Preparation Skills of Grade 9 Students in Technology and Livelihood Education

This study was conducted to determine the relationship between the distance learning home experience and the food preparation skills of Grade-9 students in Technology and Livelihood Education. The Descriptive-Correlational research design was adopted, and statistical tools such as Pearson r correlational analysis were utilized to test the relationship between the independent and dependent variables. Hypotheses were tested, and the findings revealed the following information: Using Pearson’s r correlation analysis, the data revealed that internet access is moderate negative significance to preparing fruits and vegetables, the use of a cooker, and cooking methods for Grade 9 TLE students. The knife is a moderate negative significance with online learning resources. Home assistance is moderate negative significance to preparing fruits and vegetables, using a cooker, and cooking techniques and thus the null hypothesis is not sustained. Hence, timely feedback, time allotment, and instructional delivery have no significant relationship with the food preparation skills of the learners thus the null hypothesis is sustained in this respect. It indicates that the learners experienced all the distance learning at home but when it comes to the food preparation skills, they still need the guidance of the teachers in performing their skills. It only shows that the teacher plays a vital role in improving the food preparation skills of the learners in distance learning. Having these findings, the study recommends that a comprehensive professional development program be implemented to ensure that online teachers will be given a chance to improve themselves in teaching virtually continuously; teachers are given technical assistance through training, workshop, coaching, and mentoring on class programming, and sharing of assignments, learning management specifically in managing online classes using the ICT and managing the school facilities and equipment may be considered as possible topics in the future learning and development programs for the teachers and future researchers may conduct a study on the distance learning experience of learners at home in the food preparation skills of the learners on a broader scale. Keywords: distance learning, home experience, food preparation skills, Technology, and Livelihood Education

Understanding Continued Use Intention of AI Assistants

ABSTRACT In recent years, smart home assistants have been used by a large number of people due to their simple, hands-free, voice-based operation. To ensure the long-term success and widespread dissemination of a product, it is important to evaluate its continued use. This study is mainly based on uses and gratifications theory to explore the relationship between the initial use of, gratification provided by, and continued use intention of smart home assistants, and to analyze differences in use by different age groups. The results confirm that different types of SHAs use lead to different levels of gratification in different categories. And gratification of different categories of users has a significant positive impact on the continued use intention. In addition, significant differences exist in the impact path of using smart home assistants to alleviate loneliness, among different age groups.

Complexity of Smart Home Setups: A Qualitative User Study on Smart Home Assistance and Implications on Technical Requirements

Setup and management of smart home systems is a complex task, and thus challenging for technically inexperienced users. We conducted a qualitative user study to evaluate whether an assistance system could empower users to make better and informed decisions regarding the selection of devices, their interoperability, the resulting set of features and their price. A group of 20 participants used our assistance app on a smartphone to configure a smart home while optimizing for features, interoperability, and the price-performance ratio. The results of our user study show that our assistance app can ease the problem of selecting useful devices and at the same time users become aware of new features resulting from the interoperation of selected devices. Furthermore, the assistance app can counteract the inherent interoperability problem between devices of different vendors or platforms. Finally, users are not only interested in individual device prices. They want to learn the cost of a certain feature set, including the cost of all devices necessary to realize this feature. Interestingly, none of the current smart home systems on the market offer a comparable assistance mechanism. Third-party solutions are not available either, because an assistance app requires meta data about features, interoperability, and usage of devices. This meta data is currently not available via APIs in state-of-the-art smart home systems and marketplaces. Therefore, we present a smart home architecture resulting from our research that can, among other benefits, provide the necessary meta data. Our research indicates that commercial smart home systems should invest more effort in user assistance to gain widespread adoption among technically inexperienced users. This in turn requires substantial changes to the meta data management in smart homes, because otherwise these assistance systems cannot be realized.

EXPLORING THE POTENTIAL OF VOICE INTERFACES FOR EMERGING TECHNOLOGIES: OPPORTUNITIES AND CHALLENGES

Abstract Successful human-technology interactions require that people be able to communicate their intentions to the technology. Emergent technologies such as virtual assistants, digital home assistant, smart phones, robots, and smart appliances are enabling voice activation and communication. This is potentially a positive trend as people often express a preference for using familiar methods of communicating with other people in their interactions with technologies. Presumably they expect these methods to be easier to learn and more effective. However, current voice technology interfaces are often dependent on specific language, cadence, and terminology that pose challenges for older adults to master. Moreover, age-related changes in voice characteristics, speech patterns, and the variability between individuals may lead to comprehension failures on the part of the technology. We have utilized research findings to develop implementation guidelines for use of voice interfaces with older adults. We also identify research gaps to provide a roadmap for future efforts.

ASSESSING THE POTENTIAL OF DIGITAL HOME ASSISTANTS TO SUPPORT OLDER ADULTS WITH MOBILITY DISABILITY

Abstract Digital home assistants (e.g., Amazon Echo devices) hold great potential for supporting older adults in completing a wide range of daily living activities and improving their overall quality of life. As these devices can wirelessly connect with other smart technologies to increase their capabilities (e.g., environmental control), digital assistance technologies may benefit older adults who need support within their home such as those living with mobility disabilities. However, research is needed to better understand the facilitators and barriers of using these technologies by these individuals, as well as the resources that best support their onboarding and continued use. To assess the potential efficacy of integrating smart and connected technologies into the homes of older adults with mobility disability, we are conducting a field trial to understand how best to introduce these devices and support their use. Design considerations and initial findings are discussed.