Benefits of Beamforming With Local Spatial-Cue Preservation for Speech Localization and Segregation.
Le Wang, Virginia Best, Barbara G Shinn-Cunningham
PMID 31931677WHAT IT FOUND
Adding natural two-ear spatial cues to directional microphone processing helped normal-hearing and hearing-impaired listeners localize words and understand speech among competing talkers in a lab test.
Key findings
01The full-bandwidth spatialization condition produced localization accuracy equivalent to natural binaural listening, while the beamformer without spatialization produced large localization errors.
02In the speech-on-speech task, speech reception thresholds were significantly lower in the full-bandwidth spatialization condition than in the other conditions: -12 dB for broadband speech and -11 dB for high-pass speech.
03Hearing-impaired participants had mean speech reception thresholds of -6 dB for broadband speech and -4 dB for high-pass speech, compared with -12 dB and -9 dB for normal-hearing participants, but the processing-condition effect did not differ significantly between groups.
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
Fourteen adults participated, and some analyses included only six normal-hearing participants because two could not complete all conditions. The study tested simulated processing through headphones, not hearing aids worn by patients. Participants kept their heads still, so the results may not apply to real listening with head movements. The best condition used an idealized algorithm with fine spectral and temporal resolution, and the authors note it may not be feasible in real hearing aids. The speech materials were sparse competing talkers; the authors say the method may be less effective with continuous noise or heavily overlapping sounds. Hearing-impaired participants were amplified linearly according to NAL-RP, so results may not match nonlinear hearing-aid fitting. The hearing-impaired group had losses from 2.5 to 73.3 dB HL, so the findings may not apply to every degree of hearing loss.
Declared interests
Funded by the National Institute on Deafness and Other Communication Disorders and the Air Force Office of Scientific Research. The provided text does not report conflicts of interest.
The easy way to misread this
Do not conclude that a hearing aid with this algorithm will improve your clients' speech-in-noise. The study tested a laboratory headphone simulation, not a real hearing aid, and only fixed-head listening.