Developing an IoT bathroom speaker for elderly safety
Bibliographic record
Abstract
This report presents a secure and autonomous solution for detecting falls in bathroom environments, addressing privacy concerns associated with traditional video surveillance systems. The proposed approach integrates machine learning and artificial intelligence algorithms into edge processing devices, enabling real-time decision- making at the network's edge. The system utilizes advanced audio classification models to identify conscious occupants expressing fear when calling for help, complemented by obscured thermal imaging techniques to detect unconscious fallen individuals. The audio classifier employs a Deep Neural Network (DNN) architecture trained on the Toronto Emotional Speech Data Set (TESS), achieving an overall accuracy of 88.22% in recognizing emotions from vocalizations. The thermal image classifier analyses temperature differentials between image pixels, correctly identifying fallen postures with 96% recall and 38% precision when the optimal temperature threshold is applied. Extensive testing and evaluation of the system's performance are conducted, including the construction of a thermal image dataset and the incorporation of background bathroom noise into the audio classification model, reducing the fear detection accuracy to 72.73%. The report provides a comprehensive overview of the system's methodology, hardware and software architectures, data collection and training processes, and presents the results obtained from various test scenarios. Recommendations for future work and potential enhancements are also discussed, highlighting the system's potential for widespread adoption and its contribution to enhancing elderly safety in bathroom environments while prioritizing data privacy and security.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| 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) | 0.006 | 0.004 |
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".