Speech understanding and listening effort in cochlear implant users – microphone beamformers lead to significant improvements in noisy environments

Abstract

Objectives: To evaluate the effect of microphone directionality, i.e. beamforming, on speech understanding in noise with the SONNET audio processor.

Methods: Speech reception thresholds (SRTs) were tested in three different microphone settings (omnidirectional, adaptive, and fixed beamformer (natural)) and assessed via the Oldenburg Sentence Test and the Just Understanding Speech Test. Subjects rated the listening effort needed to understand speech in different signal-to-noise ratios (−10, −5, 0, 5, 10, 15 dB SNR) via a Visual Analogue Scale. For all test methods, speech was presented at 0° azimuth while fixed and uncorrelated masking noise was presented simultaneously from five loudspeakers positioned at ±70°, ±135°, and 180° azimuth.

Results: Compared to the omnidirectional mode, significant improvements (p<0.001) were shown in mean SRTs for both the natural (3.3 dB SNR) and adaptive (5.2 dB SNR) settings. Using the natural or the adaptive setting required significantly less listening effort than using the omnidirectional setting for the SNR conditions −5 dB SNR (p=0.002) and 0 dB SNR (p<0.001).

Discussion: The beamformer settings significantly improved speech understanding in noise over the omnidirectional setting. Due our multi-speaker test setup, we conclude that beamforming should yield significantly better and less stressful speech understanding in demanding real-life listening situations.

Keywords:

• Cochlear implant
• Beamformer
• Directionality
• Audio Processor
• Speech in noise
• Listening effort

Acknowledgements

Michael Todd (MED-EL) edited the English on a version of this manuscript.