José Luis Yagüe Blanco

Using HMM to Detect Speakers with Severe Obstructive Sleep Apnoea Syndrome

Nowadays definitive diagnosis of obstructive sleep apnoea (OSA) syndrome is expensive and time-consuming. Previous research on voice characteristics of OSA patients has shown that resonance, phonation and articulation differences arise when compared to healthy subjects. In this contribution we study different speech modeling techniques to detect patients with severe OSA envisioning the future classification of patients according to their priority of need identifying the most severe cases and reducing medical costs.

Early Detection of Severe Apnoea through Voice Analysis and Automatic Speaker Recognition Techniques

This study is part of an on-going collaborative effort between the medical and the signal processing communities to promote research on applying voice analysis and Automatic Speaker Recognition techniques (ASR) for the automatic diagnosis of patients with severe obstructive sleep apnoea (OSA). Early detection of severe apnoea cases is important so that patients can receive early treatment. Effective ASR-based diagnosis could dramatically cut medical testing time. Working with a carefully designed speech database of healthy and apnoea subjects, we present and discuss the possibilities of using generative Gaussian Mixture Models (GMMs), generally used in ASR systems, to model distinctive apnoea voice characteristics (i.e. abnormal nasalization). Finally, we present experimental findings regarding the discriminative power of speaker recognition techniques applied to severe apnoea detection. We have achieved an 81.25 % correct classification rate, which is very promising and underpins the interest in this line of inquiry.

Introducing non-linear analysis into sustained speech characterization to improve sleep apnea detection

We present a novel approach for detecting severe obstructive sleep apnea (OSA) cases by introducing non-linear analysis into sustained speech characterization. The proposed scheme was designed for providing additional information into our baseline system, built on top of state-of-the-art cepstral domain modeling techniques, aiming to improve accuracy rates. This new information is lightly correlated with our previous MFCC modeling of sustained speech and uncorrelated with the information in our continuous speech modeling scheme. Tests have been performed to evaluate the improvement for our detection task, based on sustained speech as well as combined with a continuous speech classifier, resulting in a 10% relative reduction in classification for the first and a 33% relative reduction for the fused scheme. Results encourage us to consider the existence of non-linear effects on OSA patients’ voices, and to think about tools which could be used to improve short-time analysis.

SEVERE APNOEA DETECTION USING SPEAKER RECOGNITION TECHNIQUES

Proceedings of the International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS 2009)