Lorena Álvarez

Sensorized Garments and Textrode-Enabled Measurement Instrumentation for Ambulatory Assessment of the Autonomic Nervous System Response in the ATREC Project

Advances in textile materials, technology and miniaturization of electronics for measurement instrumentation has boosted the development of wearable measurement systems. In several projects sensorized garments and non-invasive instrumentation have been integrated to assess on emotional, cognitive responses as well as physical arousal and status of mental stress through the study of the autonomous nervous system. Assessing the mental state of workers under stressful conditions is critical to identify which workers are in the proper state of mind and which are not ready to undertake a mission, which might consequently risk their own life and the lives of others. The project Assessment in Real Time of the Stress in Combatants (ATREC) aims to enable real time assessment of mental stress of the Spanish Armed Forces during military activities using a wearable measurement system containing sensorized garments and textile-enabled non-invasive instrumentation. This work describes the multiparametric sensorized garments and measurement instrumentation implemented in the first phase of the project required to evaluate physiological indicators and recording candidates that can be useful for detection of mental stress. For such purpose different sensorized garments have been constructed: a textrode chest-strap system with six repositionable textrodes, a sensorized glove and an upper-arm strap. The implemented textile-enabled instrumentation contains one skin galvanometer, two temperature sensors for skin and environmental temperature and an impedance pneumographer containing a 1-channel ECG amplifier to record cardiogenic biopotentials. With such combinations of garments and non-invasive measurement devices, a multiparametric wearable measurement system has been implemented able to record the following physiological parameters: heart and respiration rate, skin galvanic response, environmental and peripheral temperature. To ensure the proper functioning of the implemented garments and devices the full series of 12 sets have been functionally tested recording cardiogenic biopotential, thoracic impedance, galvanic skin response and temperature values. The experimental results indicate that the implemented wearable measurement systems operate according to the specifications and are ready to be used for mental stress experiments, which will be executed in the coming phases of the project with dozens of healthy volunteers.

Reducing the computational cost for sound classification in hearing aids by selecting features via genetic algorithms with restricted search

This paper centers on designing a feature-selection algorithm able to provide a ldquosmallrdquo number of adequate features that assist a sound classification system for hearing aids in reducing its computational load without degrading its performance. Because of the problem complexity, we have explored the use of genetic algorithms with restricted search for the mentioned feature selection. In an effort to evaluate its performance, the algorithm has been compared to a standard unconstrained genetic algorithm and with sequential methods. The restricted search driven by the proposed algorithm performs better than both the sequential methods and unconstrained genetic algorithms. The proposed algorithm selects a feature subset composed of only 21 features, much smaller than the 76 features of the complete, original set of available features. This low-cardinality subset of signal-describing features is the one implemented on the hearing aid, saving thus a great number of the scarce computational resources, and making possible to put into practice the concept at reasonable cost.

Two-layer automatic sound classification system for conversation enhancement in hearing aids

This paper focuses on the development of an automatic sound classifier for digital hearing aids that aims to enhance the listening comprehension when the user goes from a sound environment to another different one. The approach consists in dividing the classifying algorithm into two layers that make use of two-class algorithms that work more efficiently: the input signal discriminated by the first layer into either speech or non-speech is ulteriorly classified more specifically depending on whether the user is in a conversation (both in quiet or in the presence of background noise) or in a noisy ambient in the absent of speech. The system results in having four classes, labeled speech in quiet, speech in noise, stationary noisy environments (for instance, an aircraft cabin), and non-stationary noisy environments. The combination of classifiers that has been found to be more successful in terms of probability of correct classification consists of a system that makes use of Multilayer Perceptrons for those classification tasks in which speech is involved, and a Fisher Linear Discrimnant for distinguising stationary noisy environments from the non-stationary ones. The system performance has been found to be higher than that of other more classical approaches, and even superior than that of our preliminary work.

Enhancing the energy efficiency of wireless-communicated binaural hearing aids for speech separation driven by soft-computing algorithms

Abstract: Assisted by soft computing methods, the work we present in this paper focuses on the design of energy-efficient algorithms for binaural hearing aids that aim to separate speech from other sounds the hearing impaired person is not interested in. To do this, the right and left hearing aids need to wirelessly transmit to each other some parameters involved in the speech separation algorithm. The problem is that this transmission appreciably reduces the battery life, the most important constrain for designing advanced algorithms in hearing aids. Reducing the number of bits used to represent the parameters to be transmitted will bring down the power consumption, but at the expense of degrading the ability of the system to separate the speech from the other sound sources. Aiming at solving this problem, our approach, based on quantizing the parameters to be transmitted, basically consists in computing the adequate number of quantization bits by means of a combination of neural networks and genetic algorithms in the effort of finding a balance between low bit rate (and thus, low power consumption) and good separation of speech. The results show that even by using only 2bits/quantized-sample, the quality of the separation is as high as 70% of the limiting non-quantized quality separation factor, which has been found to be 85%.

Influence of acoustic feedback on the learning strategies of neural network-based sound classifiers in digital hearing aids

Sound classifiers embedded in digital hearing aids are usually designed by using sound databases that do not include the distortions associated to the feedback that often occurs when these devices have to work at high gain and low gain margin to oscillation. The consequence is that the classifier learns inappropriate sound patterns. In this paper we explore the feasibility of using different sound databases (generated according to 18 configurations of real patients), and a variety of learning strategies for neural networks in the effort of reducing the probability of erroneous classification. The experimental work basically points out that the proposed methods assist the neural network-based classifier in reducing its error probability in more than 18%. This helps enhance the elderly users comfort: the hearing aid automatically selects, with higher success probability, the program that is best adapted to the changing acoustic environment the user is facing.

NN-based automatic sound classifier for digital hearing aids

This paper centers on exploring proper training algorithms for multilayer perceptrons (MLPs) to be used within digital hearing aids. One argument usually considered against the feasibility of neural networks on hearing aids consists in both their computational complexity and the hardware constraints the hearing aids suffer from. Within this framework, this work focuses on studying the influence of a number of training methods for an MLP able to automatic classify the sounds entering the hearing aid into three classes: speech, noise and music. The training methods explored are Gradient Descent, Levenberg-Marquardt, and Levenberg-Marquardt with Bayesian Regularization. Our results show how the proper selection of the training algorithm leads to a good mean probability of correct classification of 91.7% along with a low number of neurons, the computational complexity being thus reduced. These results have been successfully compared to those obtained from a k-Nearest Neighbors algorithm, which exhibits poorer performance.

Bioimpedance-Based Wearable Measurement Instrumentation for Studying the Autonomic Nerve System Response to Stressful Working Conditions

The assessment of mental stress on workers under hard and stressful conditions is critical to identify which workers are not ready to undertake a mission that might put in risk their own life and the life of others. The ATREC project aims to enable Real Time Assessment of Mental Stress of the Spanish Armed Forces during military activities. Integrating sensors with garments and using wearable measurement devices, the following physiological measurements were recorded: heart and respiration rate, skin galvanic response as well as peripheral temperature. The measuring garments are the following: a sensorized glove, an upper-arm strap and a repositionable textrode chest strap system with 6 textrodes. The implemented textile-enabled instrumentation contains: one skin galvanometer, two temperature sensors, for skin and environmental, and an Impedance Cardiographer/Pneumographer containing a 1 channel ECG amplifier to record cardiogenic biopotentials. The implemented wearable systems operated accordingly to the specifications and are ready to be used for the mental stress experiments that will be executed in the coming phases of the project in healthy volunteers.

Detection of Emotions and Stress through Speech Analysis

The term “stress” is referred to a psychological state, obtained in response to a situation which is accompanied by an emotional reaction, such as, for instance, anxiety, anger, sadness, etc. Emotions are psychophysiological reactions representing modes of adaptation to certain either environmental stimuli or own stimuli. They are those feelings or perceptions of the elements and relations of the reality and the imagination, which are physically expressed by some physiological functions, like, for example, facial reactions, changes in heartbeat or distortions in the paralinguistic aspects of speech. Although both stress and emotional state do not alter the linguistic content, they could affect the paralinguistic contents of speech and this is an important factor in human communication, because it provides more information about the interlocutor than the merely semantic one. This subliminal information of speech is analyzed in this paper. Throughout the present document many features and algorithms are studied, which are combined to obtain an emotion and stress detector with a high degree of reliability.

Speech Enhancement in Noisy Environments in Hearing Aids Driven by a Tailored Gain Function Based on a Gaussian Mixture Model

This paper centers on a novel approach aiming at speech enhancement in hearing aids. It consists in creating -by making use of perceptual concepts, and a supervised learning process driven by a genetic algorithm (GA)- a gain function (\(\mathcal{G}\)) that not only does it enhance the speech quality but also the speech intelligibility in noisy environments. The proposed algorithm creates the enhanced gain function by using a Gaussian mixture model fueled by the GA. To what extent the speech quality is enhanced is quantitatively measured by the algorithm itself by using a scheme based on the perceptual evaluation of speech quality (PESQ) standard. In this “blind” process, it does not use any initial information but that iteratively quantified by the PESQ measurement. The GA computes the optimized parameters that maximize the PESQ score. The experimental work, carried out over three different databases, shows how the computed gain function assists the hearing aid in enhancing speech, when compared to the values reached by using a standard hearing aid based on a multiband compressor-expander algorithm.

Synchronizing Speech Mixtures in Speech Separation Problems under Reverberant Conditions

Blind Source Separation (BSS) techniques aim at recovering unobserved source signals from observed mixtures (typically, the outputs of an array of sensors). Practically all classical BSS techniques do not work properly under reverberant conditions and therefore, it still remains an open problem. In this sense, we propose in this document the use of synchronization of speech mixtures in order to improve the results of classical BSS techniques. Specifically, we have applied the synchronization of mixtures combined with one of the most well-known and robust BSS algorithms that works under non-reverberant conditions, the Degenerate Unmixing Estimation Technique (DUET). In the aim of synchronizing speech mixtures prior to the speech source separation, the suitability of working with seven Time Delay Estimation (TDE) techniques has been analyzed. Results show the feasibility of using synchronization since the results of DUET are improved and additionally, it has been observed what is the most useful TDE algorithm in this framework.