Title: A novel method for in-vehicle speech intelligibility evaluation and statistical variability analysis

Authors: Nikolina Samardzic; Colin Novak

Addresses: Department of Mechanical, Automotive and Materials Engineering, University of Windsor, 401 Sunset Ave. Windsor, Ontario N9B 3P4, Canada ' Department of Mechanical, Automotive and Materials Engineering, University of Windsor, 401 Sunset Ave. Windsor, Ontario N9B 3P4, Canada

Abstract: A novel method for evaluating in-vehicle speech intelligibility is proposed using the speech intelligibility index (SII) based on measured speech signal, evaluated at the sentence speech reception threshold (sSRT) in a simulated driving environment. The statistical variability was quantified by considering the hearing ability of normal hearing individuals and the influence of multi-sensory perception on the in-vehicle speech intelligibility. The sSRT values were obtained from the hearing in noise test (HINT) using a jury of 30 participants in a driving simulation created by Samardzic et al. (2012). Different configurations of the talker and the listener and vehicle operating conditions were also considered. The SII values evaluated at the sSRT and the associated statistical variability presented in this study provide a benchmark for future evaluation of in-vehicle speech intelligibility.

Keywords: in-vehicle speech intelligibility; speech intelligibility evaluation; speech intelligibility index; SII; statistical variability; normal listeners; hearing impaired listeners; multi-sensory perception; hearing impairment; speech signals; simulation.

DOI: 10.1504/IJVNV.2015.070017

International Journal of Vehicle Noise and Vibration, 2015 Vol.11 No.2, pp.114 - 132

Received: 15 May 2014
Accepted: 02 Aug 2014

Published online: 24 Jun 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article