Wake-Cough: cough spotting and cougher identification for personalised long-term cough monitoring
Madhurananda Pahar, Marisa Klopper, Byron Reeve, Rob Warren, Grant, Theron, Andreas Diacon, Thomas Niesler

TL;DR
Wake-Cough is a personalized, power-efficient cough monitoring system that uses wake-word spotting and speaker identification techniques to accurately detect and identify coughers in various environments, suitable for long-term health monitoring.
Contribution
This work introduces a novel system combining wake-word detection with cougher identification using i-vectors, achieving high accuracy in both quiet and noisy settings for long-term health monitoring.
Findings
Achieves 90.02% accuracy in noisy environments for 51 coughers.
Achieves up to 99.78% accuracy with longer segments in quiet environments.
i-vectors outperform x-vectors and d-vectors in cougher identification.
Abstract
We present `wake-cough', an application of wake-word spotting to coughs using a Resnet50 and the identification of coughers using i-vectors, for the purpose of a long-term, personalised cough monitoring system. Coughs, recorded in a quiet (735 dB) and noisy (3417 dB) environment, were used to extract i-vectors, x-vectors and d-vectors, used as features to the classifiers. The system achieves 90.02\% accuracy when using an MLP to discriminate between 51 coughers using 2-sec long cough segments in the noisy environment. When discriminating between 5 and 14 coughers using longer (100 sec) segments in the quiet environment, this accuracy improves to 99.78% and 98.39% respectively. Unlike speech, i-vectors outperform x-vectors and d-vectors in identifying coughers. These coughs were added as an extra class to the Google Speech Commands dataset and features were extracted by…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRespiratory and Cough-Related Research · Speech Recognition and Synthesis · Pneumonia and Respiratory Infections
