Auditory externalization of a remote microphone signal

A remote microphone (RM) system can be used in combination with wearable binaural communication devices, such as hearing aids (HAs), to improve speech intelligibility. Typically, a speaker is equipped with a body-worn microphone which enables to pick up their voice with a high signal-to-noise ratio (SNR). However, if this signal is played diotically through the receivers (i.e. the same signal in both ears), the necessary cues enabling the auditory system to locate the sound source are bypassed. This can affect the ability of the listener to feel immersed in the environment or to follow a conversation, especially for hearing-impaired (HI) listeners. Auditory sound source localization in humans is performed by the auditory system owing to the interpretation of various cues related to the physical sound propagation from this source to the eardrums of the listener. This is mainly enabled by the binaural structure of the auditory system.Previous works have successfully developed a method to provide a simplified spatialization of the RM signal which enabled normal-hearing (NH) and HI listeners to locate sound sources in azimuth, while preserving speech intelligibility. However, the proposed method yielded common spatial hearing perceptual artefacts, such as in-head localization and front/back confusion.This thesis is devoted to the investigation and the perceptual evaluation of wearable devices-compatible audio playback solutions aiming at enhancing the realism of this spatialization. In particular, the goal is to improve the externalization of a sound source, i.e. to ensure it is perceived outside the head, and possibly at a certain distance corresponding to its physical location in the environment. For this purpose, early reflections (ERs) in a room play a key role. Hence, several signal processing approaches to provide ERs in the binaural synthesis were investigated. Subsequently, a subjective listening study was conducted, in which NH and HI aided listeners had to evaluate the perceived auditory distance with various binaural rendering strategies. The results show that the superimposition of ERs with the considered methods significantly improves the perception of auditory distance for NH and HI listeners. The study also provides insights about auditory distance perception in aided HI listeners with severe-to-profound hearing loss. A follow-up study was conducted with NH listeners and showed that a complete implementation of these strategies might improve auditory distance perception while preserving spatial awareness. In these studies, subjects had their head fixed.Previous studies have shown that head movements coupled with head-tracking can contribute to the auditory externalization of a virtual sound source. The next part of this thesis reports the development of a head-tracking algorithm compatible with wearable devices, relying solely on two 3-axis accelerometers. While showing promising results, the limitations of the developed algorithm still yield mismatches in the estimation. Consequently, a subjective listening test was conducted with NH listeners to study the effect of head-tracking artefacts on the perception of externalization and the performance in azimuth localization of a sound source. The results suggest that auditory externalization is not affected by a large latency or the amplitude mismatch. Latency did not decrease the performance in localization either, contrarily to the amplitude mismatch.

Adaptive wake and sleep detection for wearable systems

Walter Karlen

Sleep problems and disorders have a serious impact on human health and wellbeing. The rising costs for treating sleep-related chronic diseases in industrialized countries demands efficient prevention. Low-cost, wearable sleep / wake detection systems which give feedback on the wearer's "sleep performance" are a promising approach to reduce the risk of developing serious sleep disorders and fatigue. Not all bio-medical signals that are useful for sleep / wake discrimination can be easily recorded with wearable systems. Sensors often need to be placed in an obtrusive location on the body or cannot be efficiently embedded into a wearable frame. Furthermore, wearable systems have limited computational and energetic resources, which restrict the choice of sensors and algorithms for online processing and classification. Since wearable systems are used outside the laboratory, the recorded signals tend to be corrupted with additional noise that influences the precision of classification algorithms. In this thesis we present the research on a wearable sleep / wake classifier system that relies on cardiorespiratory (ECG and respiratory effort) and activity recordings and that works autonomously with minimal user interaction. This research included the selection of optimal signals and sensors, the development of a custom-tailored hardware demonstrator with embedded classification algorithms, and the realization of experiments in real-world environments for the customization and validation of the system. The processing and classification of the signals were based on Fourier transformations and artificial neural networks that are efficiently implementable into digital signal controllers. Literature analysis and empiric measurements revealed that cardiorespiratory signals are more promising for a wearable sleep / wake classification than clinically used signals such as brain potentials. The experiments conducted during this thesis showed that inter-subject differences within the recorded physiological signals make it difficult to design a sleep / wake classification model that can generalize to a group of subjects. This problem was addressed in two ways: First by adding features from another signal to the classifier, that is, measuring the behavioral quiescence during sleep using accelerometers. Conducted research on different feature extraction methods from accelerometer data showed that this data generalizes well for distinct subjects in the study group. In addition, research on user-adaptation methods was conducted. Behavioral sleep and wake measures, notably the measurement of reactivity and activity, were developed to build up a priori knowledge that was used to adapt the classification algorithm automatically to new situations. This thesis demonstrates the design and development of a low-cost, wearable hardware and embedded software for on-line sleep / wake discrimination. The proposed automatic user-adaptive classifier is advantageous compared to previously suggested classification methods that generalize over multiple subjects, because it can take changes in the wearer's physiology and sleep / wake behavior into account without adjustment from a human expert. The results of this thesis contribute to the development of smart, wearable, bio-physiological monitoring systems which require a high degree of autonomy and have only low computational resources available. We believe that the proposed sleep / wake classification system is a first promising step toward a context-aware system for sleep management, sleep disorder prevention, and reduction of fatigue.

EPFL2009

Euclidean Distance Matrices

Reza Parhizkar

Euclidean distance matrices (EDMs) are central players in many diverse fields including psychometrics, NMR spectroscopy, machine learning and sensor networks. However, they are not often exploited in signal processing. In this thesis, we analyze attributes of EDMs and derive new key properties of them. These analyses allow us to propose algorithms to approximate EDMs and provide analytic bounds on the performance of our methods. We use these techniques to suggest new solutions for several practical problems in signal processing. Together with these properties, algorithms and applications, EDMs can thus be considered as a fundamental toolbox to be used in signal processing. In more detail, we start by introducing the structure and properties of EDMs. In particular, we focus on their rank property; the rank of an EDM is at most the dimension of the set of points generating it plus 2. Using this property, we introduce the use of low rank matrix completion methods for approximating and completing noisy and partially revealed EDMs. We apply this algorithm to the problem of sensor position calibration in ultrasound tomography devices. By adapting the matrix completion framework, in addition to proposing a self calibration process for these devices, we also provide analytic bounds for the calibration error. We then study the problem of sensor localization using distance information by minimizing a non-linear cost function known as the s-stress function in the multidimensional scaling (MDS) community. We derive key properties of this cost function that can be used to reduce the search domain for finding its global minimum. We provide an efficient, low cost and distributed algorithm for minimizing this cost function for incomplete networks and noisy measurements. In randomized experiments, the proposed method converges to the global minimum of the s-stress in more than 99% of the cases. We also address the open problem of existence of non-global minimizers of the s-stress and reduce this problem to a hypothesis. If the hypothesis is true then the cost function has only global minimizers, otherwise, it has non-global minimizers. Using the rank property of EDMs and the proposed minimization algorithm for approximating them, we address an interesting and practical problem in acoustics. We show that using five microphones and one loudspeaker, we can hear the shape of a room. We reformulate this problem as finding the locations of the image sources of the loudspeaker with respect to the walls. We propose an algorithm to find these positions only using first-order echoes. We prove that the reconstruction of the room is almost surely unique. We further introduce a new algorithm for locating a microphone inside a known room using only one loudspeaker. Our experimental evaluations conducted on the EPFL campus and also in the Lausanne cathedral, confirm the robustness and accuracy of the proposed methods. By integrating further properties of EDMs into the matrix completion framework, we propose a new method for calibrating microphone arrays in a diffuse noise field. We use a specific characterization of diffuse noise fields to relate the coherence of recorded signals by two microphones to their mutual distance. As this model is not reliable for large distances between microphones, we use matrix completion coupled with other properties of EDMs to estimate these distances and calibrate the microphone array. Evaluation of our algorithm using real data measurements demonstrates, for the first time, the possibility of accurately calibrating large ad-hoc microphone arrays in a diffuse noise field. The last part of the thesis addresses a central problem in signal processing; the design of discrete-time filters (equivalently window functions) that are compact both in time and frequency. By properly adapting the definitions of compactness in the continuous time to discrete time, we formulate the search for maximally compact sequences as solving a semi-definite program. We show that the spectra of maximally compact sequences are a special class of Mathieu’s cosine functions. Using the asymptotic behavior of these functions, we provide a tight bound for the time-frequency spread of discrete-time sequences. Our analysis shows that the Heisenberg uncertainty bound on the time-frequency spread of sequences is not tight and the lower bound depends on the frequency spread, unlike in the continuous time case.

EPFL2013

Experimental evaluation of speech enhancement methods in remote microphone systems for hearing aids

Gilles André Courtois, Vincent Pierre Olivier Grimaldi, Hervé Lissek

Enhancing speech intelligibility for hearing-impaired subjects in complex acoustic conditions is still a challenging topic of research. To mitigate the detrimental effects of background noise and reverberation, current hearing instruments incorporate various hardware and software strategies, among which speech enhancement algorithms are of primary importance. In this paper, two algorithms based on the multichannel Wiener filter (previously reported in the literature) and one proprietary algorithm are experimentally assessed and compared. All of them make use of a remote microphone worn by the speaker of interest. The objective of these algorithms is to improve the speech contribution within the hearing aid microphone signals. The algorithms are assessed in terms of interference reduction performance, speech quality, spatial hearing preservation, and technical requirements. Using a recorded database of audio signals, the effects of the signal-to-noise ratio and of the delay between the remote and hearing aid microphone signals are studied. The results show that the proprietary algorithm provides a good performance and yields the lowest distortion of the binaural localization cues, while being the most efficient in terms of computational cost and wireless usage. The main drawback is the degradation of the output sound quality that is observed when the remote and hearing aid microphone signals are not temporally aligned.