8 pagesInternational audienceCirdo project is intended to establish new healthcare systems to ensure the safety at home of seniors and people with decreasing independence. In particular, extending "e-lio" (http://www.technosens.fr/) device, Cirdo aims to develop an audio/video-based system which makes it possible for elderly to live with a sufficient degree of autonomy. To achieve this goal, generic purposes on video analysis and audio processing are discussed and implemented in the same process. Audio and video analysis algorithms are launched simultaneously and thinks to GPU implementation, the tasks are done in real time. To comply with the requirements set out in the Cirdo project namely: respect privacy and preserve personal data, the processing tasks is performed through background tasks, without any human intrusion. In order to have different types of fall for our experiences, multiple scenarios were designed and performed by several persons. Besides the technical aspect, the project also focuses on generic technology validation by potential users and their human environment (family, caregivers/care takers...). Psychological and ergonomic assessment on the use of services developed was conducted. It focused on the usefulness, usability and accessibility and acceptance of the tool. This evaluation was completed by critical investigation of knowledge acquired by professionals (geriatricians, caregivers school and associations)

No abstract

No abstract

Smart homes are discussed as a win-win solution for maintaining the Elderly at home as a better alternative to care homes for dependent elderly people. Such Smart homes are characterized by rich domestic commands devoted to elderly safety and comfort. The vocal command has been identified as an efficient, well accepted, interaction way, it can be directly addressed to the "habitat", or through a robotic interface. In daily use, the challenges of vocal commands recognition are the noisy environment but moreover the reformulation and the expressive change of the strictly authorized commands. This paper focuses (1) to show, on the base of elicited corpus, that expressive speech, in particular distress speech, strongly affects generic state of the art ASR systems (20 to 30%) (2) how interesting improvement thanks to ASR adaptation can regulate (15%) this degradation. We conclude on the necessary adaptation of ASR system to expressive speech when they are designed for person's assistance.

As part of a human-robot interaction project, the gestural modality is one of many ways to communicate. In order to develop a relevant gesture recognition system associated to a smart home butler robot, our methodology is based on an IQ game-like Wizard of Oz experiment to collect spontaneous and implicitly produced gestures in an ecological context where the robot is the referee. These gestures are compared with explicitly produced gestures to determine a relevant ontology of gestures. This preliminary qualitative analysis will be the base to build a big data corpus in order to optimize acceptance of the gesture dictionary in coherence with the "socio-affective glue" dynamics.

Wearable personable emergency response (PER) systems are the mainstream solution for allowing frail and isolated individuals to call for help in an emergency. However, these devices are not well adapted to all users and are often not worn all the time, meaning they are not available when needed. This paper presents a Voice User Interface system for emergency call recognition. The interface is designed to permit hands-free interaction using natural language. Crucially, this allows a call for help to be registered without necessitating physical proximity to the system. The system is based on an ASR engine and is tested on a corpus collected to simulate realistic situations. The corpus contains French speech from 4 older adults and 13 younger people wearing an old-age simulator to hamper their mobility, vision and hearing. On-line evaluation of the preliminary system showed an emergency call error rate of 27%. Subsequent off-line experimentation improved the results (call error rate 24%), demonstrating that emergency call recognition in the home is achievable. Another contribution of this work is the corpus, which is made available for research with the hope that it will facilitate related research and quicker development of robust methods for automatic emergency call recognition in the home.

PurposeThe number of older persons in industrialized countries is steadily increasing. Seniors living alone are more numerous, and we must find solutions that will allow them to continue to stay at home comfortably and safely. Smart housings can be one of these solutions. One of the biggest challenges in ambient assisted living (AAL) is to develop smart homes that anticipate and respond to the needs of the inhabitants. Given the diverse profiles of the older adult population, it will therefore be essential to facilitate interaction with the smart home through systems that respond naturally to voice commands rather than using tactile interfaces. Method The first step in our study was to evaluate how well ambient assistive speech technology is received by the target population. We report on a user evaluation assessing acceptance and fear of this new technology. The experiment aimed at testing three important aspects of speech interaction: voice command, communication with the outside world, home automation system interrupting a person's activity. Participants were 7 older persons (71-88 years old), 7 relatives and 3 professional carers; the experiments were conducted in a smart home with a voice command using a Wizard-of-Oz technique. The second step in our study was related to the adaptation of speech recognition technologies to the older adult population. Judging by the literature this has not been extensively studied. In fact, it is known that industrialized speech recognition system models are not adapted to seniors but to other categories of the population. In order to do this we recorded a specific speech corpus (voice-age) with 7 older adults (70 to 89 years old) reading sentences (a total of 4 hours of speech). A second corpus (ERES38) of free talking (18 hours of speech) was recorded by 23 speakers (68-98 years old). These corpora were analyzed in a semi -automatic manner to reveal the aged-voice characteristics. Results & Discussion Regarding the technical aspect, it appears that some phonemes are more affected by age than others. Thus, a specific adaptation of the acoustic models for ASR is required. Regarding the acceptance aspect, voice interfaces appear to have a great potential to ease daily living for older adults and frail persons and would be better accepted than other, more intrusive, solutions. By considering still healthy and independent older persons in the user evaluation, one interesting finding was overall acceptance provided the system is not conducive to a lazy lifestyle by taking control of everything. This particular concern must be addressed in the development of smart homes that support daily living by stressing the ability to control the daily routine rather than altering it. This study shows the great interest of voice interfaces to develop efficient solution to enable the growing number of older persons to continue to live in their own homes as long as possible.

One of the biggest challenges in Ambient Assisted Living is to develop smart homes that anticipate the health needs of their inhabitants while maintaining their safety and comfort. To facilitate interactions with the smart home, systems that respond naturally to voice commands would be the most adequate for disabled and frail people. In this paper we present two studies aiming at investigating the feasibility of such interactive systems. In the first study, the acceptability of a voice interface as part of the smart home was investigated. The second study is related to the adaptation of speech recognition technologies to the senior population; population which is known to challenge standard ASR systems. To this aim, we recorded two specific speech corpora (Voice-Age and ERES38) which were analyzed in a semi automatic manner to reveal the aged-voice characteristics. Some phonemes are more affected by age than others and this study shows that, by adapting acoustic models to seniors, performances increase. Voice interfaces appear to have a great potential to ease daily living for elderly and frail persons and would be better accepted than more intrusive solutions. An interesting finding that came up is their overall acceptance provided the system does not drive them to a lazy lifestyle by taking control of everything. Smart homes must support daily living by giving seniors more ability to control rather than acting in place of people.