Detection of Talking in Respiratory Signals: A Feasibility Study Using Machine Learning and Wearable Textile-Based Sensors
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Social isolation and loneliness are major health concerns in young and older people. Traditional approaches to monitor the level of social interaction rely on self-reports. The goal of this study was to investigate if wearable textile-based sensors can be used to accurately detect if the user is talking as a future indicator of social interaction. In a laboratory study, fifteen healthy young participants were asked to talk while performing daily activities such as sitting, standing and walking. It is known that the breathing pattern differs significantly between normal and speech breathing (i.e., talking). We integrated resistive stretch sensors into wearable elastic bands, with a future integration into clothing in mind, to record the expansion and contraction of the chest and abdomen while breathing. We developed an algorithm incorporating machine learning and evaluated its performance in distinguishing between periods of talking and non-talking. In an intra-subject analysis, our algorithm detected talking with an average accuracy of 85%. The highest accuracy of 88% was achieved during sitting and the lowest accuracy of 80.6% during walking. Complete segments of talking were correctly identified with 96% accuracy. From the evaluated machine learning algorithms, the random forest classifier performed best on our dataset. We demonstrate that wearable textile-based sensors in combination with machine learning can be used to detect when the user is talking. In the future, this approach may be used as an indicator of social interaction to prevent social isolation and loneliness.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it