Recognition of Distress Calls in Distant Speech Setting: a Preliminary Experiment in a Smart Home

Abstract

This paper presents a system to recognize distress speech in the home of seniors to provide reassurance and assistance. The system is aiming at being integrated into a larger system for Ambient Assisted Living (AAL) using only one microphone with a fix position in a non-intimate room. The paper presents the details of the automatic speech recognition system which must work under distant speech condition and with expressive speech. Moreover, privacy is ensured by running the decoding on-site and not on a remote server. Furthermore the system was biased to recognize only set of sentences defined after a user study. The system has been evaluated in a smart space reproducing a typical living room where 17 participants played scenarios including falls during which they uttered distress calls. The results showed a promising error rate of 29% while emphasizing the challenges of the task.

Making Emergency Calls More Accessible to Older Adults Through a Hands-free Speech Interface in the House

“…Kaldi is an open-source state-of-the-art ASR system with a high number of tools and a strong support from the community. This new built system (Vacher et al, 2015b) was evaluated off-line using Cirdo-set in the same manner as in section 4. The challenge was to determine whether the use of SGMM-based acoustic models would improve performance through adaptation to the environment and users.…”

“…SGMMs were chosen over Deep Neural Network (DNN) models because SGMM is more adapted to situation where a low amount of adaptation data is available (Badenhorst and de Wet, 2017). This section is an extension of our previous work in (Vacher et al, 2015b).…”

La commande vocale en habitat intelligent : 15 ans d’expérience dans l’équipe GETALP

“…La figure 3.2 (à gauche) illustre la grande diversité de type de données à traiter pour réaliser cette inférence : continues (son), booléennes (état d'un périphérique), évènementielles (capteur de présence infrarouge PIR) ou discontinues (mesure de température à chaque changement). Ce système est articulé autour du séquenceur PATSH [74] illustré 3.2 (à droite) qui a permis l'intégration et l'échange de données en temps réel entre les différents modules (acquisition sonore, différenciation son/parole, Reconnaissance Automatique de la Parole ou RAP, contrôleur intelligent) et leur interaction avec le réseau domotique. L'étage Acquisition/Détection limite la durée du signal détecté 𝑠(𝑡) à 4,1 s et ne décide de la fin du signal qu'après une période de 0,25 s de silence.…”

Intelligent control system of smart home for context awareness

“…13 The mapping relationship between gestures and control commands is stored in its database, and a classifier is trained by the random forest method to recognize the gesture information perceived by the depth imaging sensor. Vacher et al 14 proposed a voice-based home control system that uses deep learning methods to recognize user speech. Mofrad and Mosse 15 presented a method to separate and recognize speech in multi-user, multi-speech contexts.…”