Speech Intelligibility and Spatial Release From Masking Improvements Using Spatial Noise Reduction Algorithms in Bimodal Cochlear Implant Users.
Ayham Zedan, Tim Jürgens, Ben Williges and 4 others
PMID 33926327WHAT IT FOUND
Bimodal cochlear implant users needed less noise to understand speech in lab tests when beamforming algorithms were active.
The binaural MVDR gave larger improvements than the monaural ADM, especially in babble. No benefit occurred when speech and noise came from the same place.
Key findings
01Both the monaural ADM and the binaural MVDR improved speech-in-noise scores by 3.5 to 7.5 dB compared with no beamforming, except when speech and noise came from the same front location.
02The MVDR gave a larger benefit than the ADM in left-side noise by 1.5 dB and in 20-talker babble by 4.0 dB.
03For ADM, moving noise away from the front to the right CI side gave 10.4 dB of improvement, compared with 7.3 dB to the left HA side, and for MVDR the values were 11.2 dB to the right and 8.7 dB to the left.
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 analysed only nine participants after one of ten was excluded, so it is a small lab experiment. All participants had the CI on the right side and the hearing aid on the left side, so the results may not transfer to users with the opposite device arrangement. The acoustic setting was a simulated cafeteria with recorded head and torso impulses, not real listening in a room with head movement. The hearing aid side was simulated with a multichannel hearing aid program and a 6 ms delay, so real hearing aid processing and individual fitting differences were not fully represented. The study measured speech reception thresholds for 50% correct words, not everyday communication, listening effort, subjective preference, or long-term device use. The noise source distances were asymmetric, with the left-side noise closer than the right-side noise, which may have affected spatial release from masking. The algorithms were research implementations, and the paper notes that commercial binaural beamformers such as StereoZoom cannot be compared directly without implementation details. The study did not evaluate subjective quality of speech.
Declared interests
The authors declared no potential conflicts of interest. Funding came from DFG Cluster of Excellence 1077 Hearing4All, DFG JU2858/2-1, the EU EFRE project VIBHear, and for author B.W. from the Medical Research Council.
The easy way to misread this
Do not read the 3.5 to 7.5 dB improvements as proof that everyday conversation becomes easier. The study measured lab speech thresholds in nine bimodal CI users in simulated cafeteria audio, and it did not assess subjective listening effort or real-world communication.