Skip to Main content Skip to Navigation

Prédiction objective de l'effet des systèmes tactiques de communication et protection sur les performances de localisation sonore

Abstract : In many civilian or military situations, hearing protection is of major importance. The listener's acoustical situational awareness must however also be preserved. Tactical Communication and Protective Systems (TCAPS) are hearing protection devices that sufficiently protect the listener's ears from hazardous sounds and preserve speech intelligibility, thus allowing low-level speech communication. However, previous studies demonstrated that TCAPS still deteriorate the listener's situational awareness, in particular the ability to locate sound sources. On the horizontal plane, this is mainly explained by the degradation of the acoustical cues normally preventing the listener from making front-back confusions. In the present PhD work, a behavioral sound localization experiment is conducted with six TCAPS: two passive and two active earplugs, and two active earmuffs. The performance in open ear condition is not retrieved with any protector, but the experiment ranks the TCAPS by type: passive earplugs lead to better performance than active earplugs, and active earmuffs induce the worst performance. As part of TCAPS development and assessment, a method predicting the protector-induced degradation of the sound localization capability, and based on electroacoustic measurements, would be more suitable than time-consuming behavioral experiments. In this context, two methods based on Head-Related Transfer Functions (HRTFs) measured on an artificial head are investigated: a template-matching model and a three-layer neural network. They are optimized to fit human sound localization performance in open ear condition. The methods are applied to the HRTFs measured with the six TCAPS, providing position-dependent localization probabilities. Compared with the behavioral results, the neural network predicts realistic performances with earplugs, but overestimates errors with earmuffs. The template-matching model predicts human performance well. However, the likelihood of the resulting probability distributions with the behavioral observations is lower than that of the neural network. Finally, both methods developed in this study are independent of the artificial head used, and can be applied to assess not only TCAPS prototypes, but also hearing aids.
Complete list of metadatas

Cited literature [176 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Friday, January 26, 2018 - 4:48:08 PM
Last modification on : Thursday, September 12, 2019 - 3:45:25 AM
Long-term archiving on: : Friday, May 25, 2018 - 1:02:45 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01694013, version 1



Thomas Joubaud. Prédiction objective de l'effet des systèmes tactiques de communication et protection sur les performances de localisation sonore. Acoustique [physics.class-ph]. Conservatoire national des arts et metiers - CNAM, 2017. Français. ⟨NNT : 2017CNAM1132⟩. ⟨tel-01694013⟩



Record views


Files downloads