Assessment of audio features for automatic cough detection

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.

[1]  Thierry Dutoit,et al.  Phase-based information for voice pathology detection , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[2]  Samy Bengio,et al.  Torch: a modular machine learning software library , 2002 .

[3]  张国亮,et al.  Comparison of Different Implementations of MFCC , 2001 .

[4]  Hiroshi Motoda,et al.  Feature Selection for Knowledge Discovery and Data Mining , 1998, The Springer International Series in Engineering and Computer Science.

[5]  Roy D. Patterson,et al.  Fixed point analysis of frequency to instantaneous frequency mapping for accurate estimation of F0 and periodicity , 1999, EUROSPEECH.

[6]  Abeer Alwan,et al.  On the interdependencies between voice quality, glottal gaps, and voice-source related acoustic measures , 2010, INTERSPEECH.

[7]  S. Birring,et al.  ERS guidelines on the assessment of cough , 2007, European Respiratory Journal.

[8]  Ron Kohavi,et al.  Feature Selection for Knowledge Discovery and Data Mining , 1998 .

[9]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[10]  Jean-Philippe Thiran,et al.  Relevant Feature Selection for Audio-Visual Speech Recognition , 2007, 2007 IEEE 9th Workshop on Multimedia Signal Processing.

[11]  Zheng Fang,et al.  Comparison of different implementations of MFCC , 2001 .

[12]  Kevin McGuinness,et al.  Chronic cough: how do cough reflex sensitivity and subjective assessments correlate with objective cough counts during ambulatory monitoring? , 2007, Thorax.