Building a Multi-Sensor Lab for Interactive Communication Research: Challenges, Workarounds, and Lessons Learned.
Ronny Kurniawan Ibrahim, Kelly Miles, Lisa Maggs and 7 others
This paper reports no patient outcomes or treatment effects — it is a detailed guide to building a multi-sensor lab for recording conversations in controlled noise.
The actual findings from using this setup are in a separate paper. Nothing here changes what you do on Monday.
Key findings
1The system captures speech, eye gaze, head and body movement, and autonomic physiology (heart rate, skin conductance, respiration, skin temperature) all at once while two people talk face-to-face in controlled background noise at four levels (60, 70, 80, 87.5 dB SPL).
2Synchronising the different recording systems was the hardest part of the build; a custom clapperboard and multiple redundant trigger signals were needed to keep all data streams time-aligned, and the authors warn that relying on fallback synchronisation adds significant manual post-processing work.
3The paper explicitly states that the empirical findings from using this lab are reported in a separate publication (Luthy et al., 2026); no statistical results, hypothesis tests, or clinical outcomes appear in this article.
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
Single lab, single implementation — not validated across sites or replicated by other groups The example shows only one session with two participants; no statistical analysis or hypothesis testing is performed The audio presentation did not include head tracking, so the simulated noise scene rotated with the listener's head rather than staying fixed in the room, reducing ecological validity All empirical findings are deferred to a separate paper; this article contains no hypothesis testing or outcome measures
Declared interests
Funded by the New South Wales Smart Sensing Network Grand Challenge Fund and the Australian Future Hearing Initiative, a collaboration under Google's Digital Future Initiative in partnership with Macquarie University, Cochlear, Next Sense, and the Shepherd Centre. The authors declared no conflicts of interest. Cochlear, a cochlear implant manufacturer, is a named partner in the funding initiative.
The easy way to misread this
Do not read this as a study of how people with hearing loss communicate in noise. It is a lab-building and engineering paper that reports no patient outcomes, no statistical comparisons, and no clinical findings. The two participants shown in the example session are not described as having hearing loss, and the actual results from using this setup are in a separate paper (Luthy et al., 2026).
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →