A Comparison of Environment Classification Among Premium Hearing Instruments.
Anusha Yellamsetty, Erol J Ozmeral, Robert A Budinsky and 1 others
PMID 33749410WHAT IT FOUND
Premium hearing aids classify environments very differently from each other and from human listeners.
Devices A, B, and E agreed with humans on speech in quiet, but C and D switched to noise classes as volume rose. Music scenes caused the most disagreement across all brands.
Key findings
01In speech-in-quiet scenes, devices A, B, and E matched human judgments, while devices C and D shifted to 'speech in noise' classifications as overall level increased above 70 dB SPL.
02No devices mirrored human judgments for music presented in noise; humans heard music as the foreground roughly half the time, while devices mostly classified the scene as noise or speech in noise.
03Device C consistently classified all tested scenes as 'speech in noise' regardless of signal-to-noise ratio or background, unlike other devices which varied their output based on acoustic conditions.
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 normal-hearing listeners as the human reference group, so results may not reflect how patients with hearing loss perceive these environments. The acoustic scenes were static and did not include moving sound sources or reverberation, which are common in real-world listening. The study did not evaluate whether the classifiers' decisions were 'wrong' or intentionally biased to support specific signal processing goals set by the manufacturers. Device programming was limited to default 'first fit' settings for a mild hearing loss; custom fittings might yield different classification behaviors.
Declared interests
The authors declared no potential conflicts of interest. The study was supported by non-U.S. government funding.
The easy way to misread this
Do not interpret the differences between human judgments and device classifications as evidence that the hearing aids are malfunctioning or inaccurate. The authors note that manufacturers may intentionally bias classifiers toward certain categories to trigger specific signal processing features, such as noise reduction, even if that does not match the listener's perception of the foreground.