Andressa Ferreira State University Of Ceara andressa.magda@aluno.uece.br Adson Damasceno State University Of Ceara adson.damasceno@aluno.uece.br Mariela Cortés State University Of Ceara mariela@larces.uece.br Fran Oliveira State University Of Ceara fran.oliveira@uece.br ABSTRACT With increasing of life expectancy innovative solutions that ensure well-being of the seniors become most needed. In this context, in order to deny the main difculties reported by the senior public with the use of mobile devices, we propose a technology entirely based on voice interactions and we name this project of Guardian. The essence of the project is to provide, through a mobile application, a platform with a set of intelligent agents focused on the well-being of the older adults. In this work the mains objectives to analyze and evaluate the usability of the Guardian and the cultural impacts of the technology. For the data collection, video recordings were used, a questionnaire that identifes the socio-technological profle of the research participants and the system usability scale (SUS). KEYWORDS Elderly, Voice interaction, Conversational agent, Natural language processing, Mobile application 1 INTRODUCTION Following a worldwide trend, in Brazil, demographic data has been showing an increase in the proportion of the number of older adults over the total population. Projections by the Brazilian Institute of Geography and Statistics (IBGE) estimate that, by 2060, one in four Brazilians will be over 65 years [1]. With increasing life expectancy, longevity has come to be seen not only as a gain, but also as a concern for society. According to Nahas [2], aging human population leads to an increase in the incidence of chronic-degenerative diseases, more common in old age. This leads to a growing contingent of people who, sooner or later, become dependent on others for even the simplest tasks, such as going to the bathroom or making a cell phone call. This accelerated growth of the elderly population represents major challenges not only for families but also for society, mainly due to the decrease in the number of health professionals compared to the number of elderly people. Caregivers often complain of over- load, depression, stress, and anxiety. They leave aside, frequently, the profession, the leisure activities and self-care, which can lead to impairments in the quality of life and care provided to the elderly [3ś5]. In this context, the use of smart mobile assistants becomes a promising solution to alleviate the problem of a greater demand for care services for the elderly, contributing not only to helping the activities of caregivers and family members, but also in the recovery of the elderly independence. With this scenario, this work positions itself in a reality where there is a shortage of computational tools to promote the autonomy and the empowerment of the elderly. Empowerment for the elderly provides them with an expanded view of power, to know how to interact with society, and know how to handle and confront with their everyday situations. This empowerment can be ofered to them through methods that can be developed in a particular or collective way stimulating their self-sufciency. In addition, it is also known that an independent life reduces the cost to society of supporting people who have lost some autonomy [6]. Our project, Guardian, aims to provide a friendly technology, ac- cessible and entirely based on voice interactions, aiming to provide and facilitate the use, for the elderly who don’t have a caregiver or those who wish to live in a more autonomous way, a platform in which the same has access to diversifed and interesting services, services directed from their personal care to their entertainment. The Guardian technology consists of a set of conversational assistants focused on helping healthy aging and available in a single application embedded on a mobile device. Unlike Amazon’s, Apple’s and Google’s general propose conversational plataforms, in our approach, there is only one application running on the equipment. With this sole propose, it is expected that problems associated with Discoverability Ð common when interacting with conversational agents [7] Ð to be lessened. This multi-agent architecture was implemented using a variety of technologies, such as a Google tool called Dialogfow and Android Native. Our technology has the diferential of providing more than care, provide a friend, an individual who is fully available and that gradually learns and creates an intimate relationship with its user. Our goal is a continuous adherence of technology in the daily life of the elderly who wish to live more autonomously of caregivers and family members or simply don’t have access to such care. It is worth mentioning that a search conducted in academic libraries for similar solutions didn’t produce signifcant results. We look for mobile platforms of intelligent assistants targeted at the elderly public. We use the descriptors for the elderly, conversation assistants, mobile and synonyms of the same. This made us realize that this is an area still under-explored and motivated us to continue with the project. However, given the innovative content of the proposal there are several questions to be explored. Thus, this present study was designed to address the following research questions: 1) Are per- sonal assistants like Amazon’s Alexa or Google’s assistant ofer any help for third world country seniors? 2) How does such popula- tion receive and interact with such technology? 3) Which culture related issues might arise from the introduction of the technology in their daily lives? 4) Which possible positive impacts can be fore- Conversational Agents for the Elderly, the Guardian Platform