Assessment of Audio Features for Automatic Cough Detection

TL;DR

This study evaluates various audio features for automatic cough detection to aid respiratory disease diagnosis, analyzing feature relevance and classifier performance to improve detection accuracy.

Contribution

It introduces a comprehensive set of audio features and assesses their effectiveness with multiple classifiers for cough detection.

Findings

Features with high mutual information improve detection accuracy.

Classifier choice significantly impacts performance.

Feature dimensionality influences classifier efficiency.

Abstract

This paper addresses the issue of cough detection using only audio recordings, with the ultimate goal of quantifying and qualifying the degree of pathology for patients suffering from respiratory diseases, notably mucoviscidosis. A large set of audio features describing various aspects of the audio signal is proposed. These features are assessed in two steps. First, their intrisic potential and redundancy are evaluated using mutual information-based measures. Secondly, their efficiency is confirmed relying on three classifiers: Artificial Neural Network, Gaussian Mixture Model and Support Vector Machine. The influence of both the feature dimension and the classifier complexity are also investigated.