PLoS ONE (Jan 2020)

Improved binaural speech reception thresholds through small symmetrical separation of speech and noise.

  • Luise Wagner,
  • Lukas Geiling,
  • Christopher Hauth,
  • Thomas Hocke,
  • Stefan Plontke,
  • Torsten Rahne

DOI
https://doi.org/10.1371/journal.pone.0236469
Journal volume & issue
Vol. 15, no. 8
p. e0236469

Abstract

Read online

Speech perception in noise is challenging and is improved by binaural hearing. Since signal processing of assistive hearing devices often modifies or masks the peripheral binaural head-shadow or better-ear effects, central binaural processing should be measured separately. In a prospective study, 10 listeners with normal hearing were tested with the German matrix sentence test in a set-up with two loudspeakers located at opposite angles in the horizontal plane with respect to S0N0. The speech reception threshold (SRT) was investigated depending on the separation angle between speech and noise. The lowest (best) SRT was obtained for a separation of target and interfering source from S0N0 at an angle of about S±60°N∓60°. The derived normative curve was comparable to SRTs predicted by the binaural-speech-intelligibility-model. The systematic separation of signal and noise showed a significant improvement in speech intelligibility for normal-hearing people even for small separation angles. This experimental setting was verified. This study aimed to assess the effect of small sound source separation on binaural hearing and speech perception.