Speech recognition with a hearing-aid processing scheme combining beamforming with mask-informed speech enhancement.
Tim Green, Gaston Hilkhuysen, Mark Huckvale and 6 others
PMID 34985356WHAT IT FOUND
Binaural beamforming improved speech recognition in noise more than bilateral beamforming for hearing-impaired listeners.
Adding mask-informed speech enhancement to beamforming did not help and often slightly worsened performance compared to beamforming alone.
Key findings
01Binaural beamforming provided greater benefit to sentence recognition than bilateral beamforming for hearing-impaired listeners in a simulated classroom.
02Combining beamformers with mask-informed speech enhancement did not improve performance and often made it slightly poorer than beamforming alone.
03For hearing-impaired listeners, the benefit of binaural over bilateral beamforming averaged around 1 to 2 dB in speech reception threshold.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study used simulated environments rather than real-world testing. The beamformers were optimally adapted for the target source, which may not reflect dynamic real-world listening conditions where the target moves. The mask-informed enhancement used a deep neural network trained on specific noise types, which may not generalize to all real-world noise backgrounds. The study did not assess the impact of head movements or visual cues, which are common in real-world communication.
Declared interests
The authors declared no potential conflicts of interest. The work was supported by a grant from the Engineering and Physical Sciences Research Council.
The easy way to misread this
Do not conclude that all speech enhancement algorithms are ineffective. The study only tested one specific type of mask-informed enhancement combined with beamforming. Other enhancement methods like Wiener filtering and Kalman filtering, tested only in normal-hearing listeners, performed better than the mask-informed technique but still did not improve upon beamforming alone.