Development of a Method to Assess In-Ear Speech Intelligibility Through Listening Effort
Bibliographic record
Abstract
Active hearing protection devices equipped with an in-ear microphone enable in-ear voice pickup which, presents better signal-to-noise ratio over ambient microphones in highly noisy conditions. To improve its intelligibility, in-ear speech requires processing which can take many forms, from fixed filtering to spectral domain processing based on machine learning. Comparing these processing strategies objectively can be difficult. In the authors’ experience, objective intelligibility assessment techniques like the modified rhyme test have not provided the necessary resolution to compare various processed in-ear speech. They have also failed to capture a concept of listening effort, which intuitively seemed to increase with in-ear speech over reference speech whenrecorded in silence.This work presents the early development and preliminary validation of a technique that objectively measures intelligibility of speech material and its associated listening effort. Using a dual task paradigm, the attentional and cognitive resources required to understand speech were quantified. Ten participants performed a closed-set word recognition task and visual pattern recognition task separately and concurrently. Accuracy data were collected for in-ear and reference speech in noise. Dual-task costs (DTC) were calculated.Speech intelligibility was low and listening effort was high for reference speech presented with an 85 dB(A) competing noise. When reducing the competing noise level to 80 dB(A), speech intelligibility improved, but listening effort remained important. While in-ear speech (with and without speech processing) presented with high levels of intelligibility, they also presented with better listening effort than reference speech with competing noise. Our results suggest that a dual-task paradigm to measure listening effort can be a good approach to behaviorally evaluate in-ear speech. Our pilot study supports the use of in-ear speech to improve communication in noisy settings while quantifying its potential for further improvements.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".