Binaural Speech Intelligibility in Noise and Reverberation: Prediction of Group Performance for Normal-hearing and Hearing-impaired Listeners.
Raphael Cueille, Mathieu Lavandier
PMID 40432370WHAT IT FOUND
A new computer model accurately predicts how hearing loss and reverberation affect speech understanding in noise.
It combines two existing approaches to handle complex listening conditions that neither previous model could solve alone, offering a better tool for assessing real-world communication difficulties.
Key findings
01The new model accurately predicts speech reception thresholds for both normal-hearing and hearing-impaired listeners in conditions involving reverberation, modulated noise, and spatial separation.
02Two versions of the model, one using a fixed early-late limit and one based on physical mixing time, provided the most accurate predictions across all tested datasets.
03The model currently cannot account for all aspects of hearing loss, such as reduced compression or broadened auditory filters, which limits its ability to predict individual variability among hearing-impaired listeners.
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 model requires detailed input data (raw signals and room impulse responses) that are not readily available in typical clinical settings, limiting its immediate use for patient-specific assessments. It does not account for several factors affecting hearing loss, such as reduced compression, broadened auditory filters, or age-related changes, which explains why it predicts group trends well but struggles with individual variability. The model slightly overestimates the benefit of listening in the 'dips' of modulated noise. It cannot predict situations where late reflections of speech are actually helpful for intelligibility, which is a rare but possible condition.
Declared interests
The authors received no financial support for the research. The work was conducted within the LabEx CeLyA and received funding from the European Union’s Horizon Europe research and innovation programme.
The easy way to misread this
Do not assume this model can predict how a specific individual patient will perform in a specific room. It accurately predicts group averages, but it misses key physiological aspects of hearing loss that cause individual variability, and it requires technical data not available in standard clinical practice.