Speech analysis for Ambient Assisted Living : technical and user design of a vocal order system

International audienceEvolution of ICT led to the emergence of smart home. A Smart Home consists in a home equipped with data-processing technology which anticipates the needs of its inhabitant while trying to maintain their comfort and their safety by action on the house and by implementing connections with the outside world. Therefore, smart homes equipped with ambient intelligence technology constitute a promising direction to enable the growing number of elderly to continue to live in their own homes as long as possible. However, the technological solutions requested by this part of the population have to suit their specific needs and capabilities. It is obvious that these Smart Houses tend to be equipped with devices whose interfaces are increasingly complex and become difficult to control by the user. The people the most likely to benefit from these new technologies are the people in loss of autonomy such as the disabled people or the elderly which cognitive deficiencies (Alzheimer). Moreover, these people are the less capable of using the complex interfaces due to their handicap or their lack ICT understanding. Thus, it becomes essential to facilitate the daily life and the access to the whole home automation system through the smart home. The usual tactile interfaces should be supplemented by accessible interfaces, in particular, thanks to a system reactive to the voice ; these interfaces are also useful when the person cannot move easily. Vocal orders will allow the following functionality: - To ensure an assistance by a traditional or vocal order. - To set up a indirect order regulation for a better energy management. - To reinforce the link with the relatives by the integration of interfaces dedicated and adapted to the person in loss of autonomy. - To ensure more safety by detection of distress situations and when someone is breaking in the house. This chapter will describe the different steps which are needed for the conception of an audio ambient system. The first step is related to the acceptability and the objection aspects by the end users and we will report a user evaluation assessing the acceptance and the fear of this new technology. The experience aimed at testing three important aspects of speech interaction: voice command, communication with the outside world, home automation system interrupting a person's activity. The experiment was conducted in a smart home with a voice command using a Wizard of OZ technique and gave information of great interest. The second step is related to a general presentation of the audio sensing technology for ambient assisted living. Different aspect of sound and speech processing will be developed. The applications and challenges will be presented. The third step is related to speech recognition in the home environment. Automatic Speech Recognition systems (ASR) have reached good performances with close talking microphones (e.g., head-set), but the performances decrease significantly as soon as the microphone is moved away from the mouth of the speaker (e.g., when the microphone is set in the ceiling). This deterioration is due to a broad variety of effects including reverberation and presence of undetermined background noise such as TV radio and, devices. This part will present a system of vocal order recognition in distant speech context. This system was evaluated in a dedicated flat thanks to some experiments. This chapter will then conclude with a discussion on the interest of the speech modality concerning the Ambient Assisted Living