Speech Modifications to Increase the Intelligibility of Vocal Messages Broadcast by Driving Assistance Systems Intended For Hearing-Impaired Drivers - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Acta Acustica united with Acustica Année : 2018

Speech Modifications to Increase the Intelligibility of Vocal Messages Broadcast by Driving Assistance Systems Intended For Hearing-Impaired Drivers

Résumé

The purpose of the AIDA(Automobile Intelligible pour Déficients Auditifs-Intelligible automobile for hearing impaired)project is to develop speech modifications tools to increase the intelligibility of voice messages broadcast by driving assistance systems embedded in personal vehicles. The target audience for this project is people suffering from mild and moderate hearing loss. This article presents an approach to increase the intelligibility of natural speech signals by signal processing methods based on filtering and dynamic compression. The magnitude of the filter wasdefined from aglobal SII (Speech Intelligibility Index) optimisation method in adriving situation incorporating hearing loss via the use of ahearing loss simulator.The effect of these speech modifications on in-telligibility levelwas assessed using aword test on ten hearing-impaired subjects. The results showasignificant increase in the speech reception threshold (SRT)i nt wo different in-car noises. The SRTw as shifted by about −9d Bt o− 11 dB depending on the in-car noise compared to unmodified words with identical sound pressure levels.
Fichier principal
Vignette du fichier
article_insa_V10.pdf (488.93 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01912756 , version 1 (12-11-2018)

Identifiants

Citer

Jérémy Rouch, Etienne Parizet. Speech Modifications to Increase the Intelligibility of Vocal Messages Broadcast by Driving Assistance Systems Intended For Hearing-Impaired Drivers. Acta Acustica united with Acustica, 2018, 104, pp.668 - 677. ⟨10.3813/AAA.919206⟩. ⟨hal-01912756⟩

Collections

INSA-GROUPE ANR
36 Consultations
128 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More