Deep Learning Features for Robust Detection of Acoustic Events in Sleep-Disordered Breathing

TL;DR

This paper introduces a novel deep learning-based acoustic feature extraction method for detecting sleep-disordered breathing sounds, demonstrating improved accuracy over traditional features and exploring system architectures with language models.

Contribution

It proposes a new bottleneck feature extraction technique using autoencoders on auditory data, and compares tandem and hybrid detection systems with language models for SDB sound identification.

Findings

Bottleneck features outperform traditional MFCCs in detection accuracy.

Tandem system outperforms hybrid system with limited training data.

Language models provide a small but consistent performance improvement.

Abstract

Sleep-disordered breathing (SDB) is a serious and prevalent condition, and acoustic analysis via consumer devices (e.g. smartphones) offers a low-cost solution to screening for it. We present a novel approach for the acoustic identification of SDB sounds, such as snoring, using bottleneck features learned from a corpus of whole-night sound recordings. Two types of bottleneck features are described, obtained by applying a deep autoencoder to the output of an auditory model or a short-term autocorrelation analysis. We investigate two architectures for snore sound detection: a tandem system and a hybrid system. In both cases, a `language model' (LM) was incorporated to exploit information about the sequence of different SDB events. Our results show that the proposed bottleneck features give better performance than conventional mel-frequency cepstral coefficients, and that the tandem system outperforms the hybrid system given the limited amount of labelled training data available. The LM made a small improvement to the performance of both classifiers.