Abstract
AbstractThis article describes the construction phase of a corpus of everyday life sounds for a system of separation and classification of audio sources in a habitat: an application for telemonitoring of the elderly or disabled. We first present the key concepts of the research area, and then we present works and projects that address detection of distress situations and recognition of activities. Thereafter, we discuss the different sounds used in an application for telemonitoring, which inspired us to create our database. Finally, we present our experiment aimed at classifying the various environmental sounds in a habitat. The objective of this experiment is to present the global idea of our future work and to explain the necessity of creating a database of everyday life sounds.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have