Neurodegenerative motor disorders affect the neuromuscular system challenging daily life and normal activity. Parkinson’s Disease (PD) is among the most prevalent ones, with a large impact and rising prevalence rates. Speech is most affected by PD as far as phonatory and articulatory performance is concerned. Neuromotor activity (NMA) alterations have an impact on larynx muscles responsible for vocal fold adduction and abduction, hampering phonation stability and regularity. The main muscular articulators involved in phonation control are the cricothyroid (tensor) and thyroarytenoid (relaxer) systems, regulated by two distinct direct neuromotor pathways, activated by the precentral gyrus laryngeal control areas. These articulations control the musculus vocalis, directly responsible for regular vocal fold vibration. An indirect estimation of the muscular tension produced by inverse filtering may split into two independent channels, assumed to be the tensor and relaxer neuromotor pathways such as the differential neuromotor activity (DNMA). The amplitude distributions of both DNMA channels allow comparing phonations from PD-affected persons (PDPs) and age-matched healthy control participants (HCPs) with respect to a set of reference mid-age normative participants (RSPs). The comparisons are carried out by Jensen–Shannon distributions of PDP and HCP phonations with respect to those of RSPs. A dataset of 96 phonation samples from participants balanced by gender is used to train a set of decision tree classifiers (DTCs) to distinguish PDP from HCP phonation. The best results from 10-fold cross-validation offered accumulated mismatches of 0.09 and 0.1292 for male and female subsets. The sensitivity, specificity, and accuracy of the classification results when separating PDP from HCP phonatios were 93.33%, 88.23%, and 90.63% (male PDP versus HCP) and 92.86%, 83.33%, and 87.50% (female PDP versus HCP), providing a stratification of PDPs and HCPs by objective disease grading from explainable AI (XAI) methods.

Parkinson's disease (PD) is a neurodegenerative condition with constantly increasing prevalence rates, affecting strongly life quality in terms of neuromotor and cognitive performance. PD symptoms include voice and speech alterations, known as hypokinetic dysarthria (HD). Unstable phonation is one of the manifestations of HD. Repetitive transcranial magnetic stimulation (rTMS) is a rehabilitative treatment thathas been shown to improve some motor and non-motor symptoms of persons with PD (PwP). This study analyzed the phonation functional behavior of 18 participants (13 males, 5 females) with PD diagnosis before (one pre-stimulus) and after (four post-stimulus) evaluation sessions of rTMS treatment, to assess the extent of changes in their phonation stability. Participants were randomized 1:1 to receive either rTMS or sham stimulation. Voice recordings of a sustained vowel [a:] taken immediately before and after the treatment, and at follow-up evaluation sessions (immediately after, at six, ten, and fourteen weeks after the baseline assessment) were processed by inverse filtering to estimate a biomechanical correlate of vocal fold tension. This estimate was further band-pass filtered into EEG-related frequency bands. Log-likelihood ratios (LLRs) between pre- and post-stimulus amplitude distributions of each frequency band showed significant differences in five cases actively stimulated. Seven cases submitted to the sham protocol did not show relevant improvements in phonation instability. Conversely, four active cases did not show phonation improvements, whereas two sham cases did. The study provides early preliminary insights into the capability of phonation quality assessment by monitoring neuromechanical activity from acoustic signals in frequency bands aligned with EEG ones.

Speech is controlled by axial neuromotor systems, therefore, it is highly sensitive to the effects of neurodegenerative illnesses such as Parkinson's Disease (PD). Patients suffering from PD present important alterations in speech, which are manifested in phonation, articulation, prosody, and fluency. These alterations may be evaluated using statistical methods on features obtained from glottal, spectral, cepstral, or fractal descriptions of speech. This work introduces an evaluation paradigm based on Information Theory (IT) to differentiate the effects of PD and aging on glottal amplitude distributions. The study is conducted on a database including 48 PD patients (24 males, 24 females), 48 age-matched healthy controls (HC, 24 males, 24 females), and 48 mid-age normative subjects (NS, 24 males, 24 females). It may be concluded from the study that Hierarchical Clustering (HiCl) methods produce a clear separation between the phonation of PD patients from NS subjects (accuracy of 89.6% for both male and female subsets), but the separation between PD patients and HC subjects is less efficient (accuracy of 75.0% for the male subset and 70.8% for the female subset). Conversely, using feature selection and Support Vector Machine (SVM) classification, the differentiation between PD and HC is substantially improved (accuracy of 94.8% for the male subset and 92.8% for the female subset). This improvement was mainly boosted by feature selection, at a cost of information and generalization losses. The results point to the possibility that speech deterioration may affect HC phonation with aging, reducing its difference to PD phonation.

Neurodegenerative pathologies as Parkinson's Disease (PD) show important distortions in speech, affecting fluency, prosody, articulation and phonation. Classically, measurements based on articulation gestures altering formant positions, as the Vocal Space Area (VSA) or the Formant Centralization Ratio (FCR) have been proposed to measure speech distortion, but these markers are based mainly on static positions of sustained vowels. The present study introduces a measurement based on the mutual information distance among probability density functions of kinematic correlates derived from formant dynamics. An absolute kinematic velocity associated to the position of the jaw and tongue articulation gestures is estimated and modeled statistically. The distribution of this feature may differentiate PD patients from normative speakers during sustained vowel emission. The study is based on a limited database of 53 male PD patients, contrasted to a very selected and stable set of eight normative speakers. In this sense, distances based on Kullback-Leibler divergence seem to be sensitive to PD articulation instability. Correlation studies show statistically relevant relationship between information contents based on articulation instability to certain motor and nonmotor clinical scores, such as freezing of gait, or sleep disorders. Remarkably, one of the statistically relevant correlations point out to the time interval passed since the first diagnostic. These results stress the need of defining scoring scales specifically designed for speech disability estimation and monitoring methodologies in degenerative diseases of neuromotor origin.

Speech articulation is produced by the movements of muscles in the larynx, pharynx, mouth and face. Therefore speech shows acoustic features as formants which are directly related with neuromotor actions of these muscles. The first two formants are strongly related with jaw and tongue muscular activity. Speech can be used as a simple and ubiquitous signal, easy to record and process, either locally or on e-Health platforms. This fact may open a wide set of applications in the study of functional grading and monitoring neurodegenerative diseases. A relevant question, in this sense, is how far speech correlates and neuromotor actions are related. This preliminary study is intended to find answers to this question by using surface electromyographic recordings on the masseter and the acoustic kinematics related with the first formant. It is shown in the study that relevant correlations can be found among the surface electromyographic activity (dynamic muscle behavior) and the positions and first derivatives of the first formant (kinematic variables related to vertical velocity and acceleration of the joint jaw and tongue biomechanical system). As an application example, it is shown that the probability density function associated to these kinematic variables is more sensitive than classical features as Vowel Space Area (VSA) or Formant Centralization Ratio (FCR) in characterizing neuromotor degeneration in Parkinson's Disease.

Abstract Most of the parameters proposed for the characterization of the emotion in speech concentrate their attention on phonetic and prosodic features. Our approach goes beyond trying to relate the biometrical signature of voice with a possible neural activity that might generate alterations in voice production. A total of 68, acoustical, glottal and biomechanical parameters were extracted from neutral and stressed speeches. The importance of the parameters was evaluated using t‐test, entropy, Receiver Operator Characteristic (ROC) and Wilcoxon methods and support vector machines algorithms for classification. The emotion under study is the stress produced when a speaker has to defend an idea opposite to his/her thoughts or feelings, and this stress is compared to self‐consistent speech. The results show tremor in the vocal folds to be the most relevant feature.

Oligonucleotide Microarrays have become powerful tools in Genetic Research, as they serve as parallel scanning mechanisms to detect the presence of genes using test probes composed of controlled segments of gene code built by masking techniques. The

MOTIVATION: Modern machine learning methods based on matrix decomposition techniques, like independent component analysis (ICA) or non-negative matrix factorization (NMF), provide new and efficient analysis tools which are currently explored to analyze gene expression profiles. These exploratory feature extraction techniques yield expression modes (ICA) or metagenes (NMF). These extracted features are considered indicative of underlying regulatory processes. They can as well be applied to the classification of gene expression datasets by grouping samples into different categories for diagnostic purposes or group genes into functional categories for further investigation of related metabolic pathways and regulatory networks. RESULTS: In this study we focus on unsupervised matrix factorization techniques and apply ICA and sparse NMF to microarray datasets. The latter monitor the gene expression levels of human peripheral blood cells during differentiation from monocytes to macrophages. We show that these tools are able to identify relevant signatures in the deduced component matrices and extract informative sets of marker genes from these gene expression profiles. The methods rely on the joint discriminative power of a set of marker genes rather than on single marker genes. With these sets of marker genes, corroborated by leave-one-out or random forest cross-validation, the datasets could easily be classified into related diagnostic categories. The latter correspond to either monocytes versus macrophages or healthy vs Niemann Pick C disease patients.

The parameterization of vocal cord dynamics may be useful in different fields of voice processing, as in speech synthesis, voice pathology detection, or speaker identification, among others. A technique to determine biomechanical cord parameters is presented to produce estimates of mass, stiffness and losses from the body and cover structures of the vocal cords. Results from a database of normophonic subjects are in good agreement with values used in cord modelling. This technique provides biometrical descriptors of speakers with use in different fields of application

In recent years emphasis has been placed upon the early detection of voice pathologies by using the signal processing of voice to evaluate certain time and spectrum domain parameters which may infer the presence of pathology. The present work is aimed at establishing the suitability of these voice spectral parameters in fixing a clear distinction between pathologic and normophonic voice, and to further classify the specific patient's pathology. Principal component analysis is used in parameter selection. Results for normal and pathological samples will be presented and discussed.

Pathology detection methods in vocal cords using the voice may benefit from the detection of the time and frequency patterns of the mucosal wave correlate. A description of the vocal folds is given in terms of the most relevant features related to vocal fold pathology in reference to the body-cover models of 2- and k-masses. A description of the mucosal wave correlate as the residual small-amplitude vibration when removing the body influence in the glottal aperture is given. The main features of the mucosal wave correlate are examined. Finally, a parameterization method based on this study and its possible applications to speech pathology detection are discussed.

The impact of voice pathologies caused by physiological alterations of the vocal cords is becoming a very important issue due to vocal abuse and unhealthy habits. Early detection of incipient damage to the cords may help in improving the prognosis, treatment and care of these pathologies. Information derived from the speech signal may help in detecting early stages of pathology, and to prevent them by assisting experts in voice therapy to correct vocal abuse in children and in professionals depending on voice as, for example, speakers, singers or lecturers. The present paper is devoted to detecting the presence of certain pathologies in the voice or speech signal from the cepstral contents of the mucosal wave reconstructed by inverse filtering, using conclusions derived from the behavior of a 2-mass vocal cord model. This can be of application in speech and singing education, and in pathology screening.

Current voice recognition systems tend to be implemented as embedded proprietary solutions. This model is not suitable for the growing complexities of present and future developments: It is single-user, it is non portable, and it assumes the workstation model, where all the CPU resources are supposed to he locally available. This work researches how a high performance speech recognition system can be redesigned and implemented as a time-critical network service with three main design goals: Scalability, predictability and POSIX portability. The whole idea has been tested by rebuilding IVORY, a well known robust desktop voice recognition methodology, as a distributed service. Also, potential fields of application are identified.

Abstract Current voice recognition systems tend to be implemented asa PC desktop facility. This model is not suitable for thegrowing complexities of present and future developments: Itis single-user, it is non portable, and it assumes theworkstation model, where all the CPU resources are supposedto be locally available. This work researches how a highperformance speech recognition system can be redesigned andimplemented as a time-critical network service shared throughordinary data transmission media with three main designgoals: Scalability, predictability and POSIX portability. Thewhole idea has been tested by rebuilding IVORY, a wellknown robust desktop voice recognition methodology, as adistributed component. 1. Introduction While Speech Processing and Recognition is a fieldexperiencing a rapid and promising expansion, the operating-system environments for the desktop PC still typically lack oftrue real-time support. To overcome this limitation, currentspeech recognition systems are confident on the workstationprinciple: all the CPU resources are always available to theapplication where they are embedded. This approach shows amain limitation: Its growing computational complexity.IVORY ([1], [7]), a stand-alone speech recognition system ofisolated words, gives figures of computational complexityaround 21 Mflop/s. Though this load is easily assumed bycurrent CPU's, continuous speech can raise the computingpower demand one order of magnitude. Noise cancellationdemands up to five or six times the power of the recognitionitself. Furthermore, new applications of speech processingdemand much more computing power. For instance, tracking asingle speaker by the Microphone Arrays technique shows acomputational complexity near 166 Mflop/s ([6]). Thoughtoday's PC microprocessors claim peak execution ratesexceeding 1 Gflop/s, regular DSP algorithms rarely result insuch a high performance. In our view, desktop speechprocessing is -and will always be- strongly limited by itscomputational complexity, nowadays constrained to thecomputing power of the average personal computer. This work was founded by CICYT and Junta de Extremaduraunder the TIC99-0609 (DIARCA) and CICYTEX IPRR98A039 projects respectively.Distributed computing should change this scenery.Ongoing developments on component based softwareengineering makes possible to envision a remote service ofDSP computing power for speech processing. It would allowto bring both to the current desktop PC and to the futureinternet appliances the more advanced developments on thefield. This work investigates the distribution of speechrecognition in the context of DIARCA, a research projectwhose aim is two-fold. Firstly, to distribute IVORY with threedesign goals: Scalability, predictability and POSIX portability.Secondly, to extend the results in order to support microphonearray developments. This work is about the first goal.Figure 1. IVORY: a Robust Voice Recognition system

Abstract The localization of acoustic sources in a room is essential in many applications, as security monitoring, video conferencing, automatic scene analysis [6], reverberation canceling [5], or robust Speech Recognition under multiple-party effect [7]. Through the present paper the design and operation of a negative beamformer for multiple source speech separation will be presented. The problems found for its proper operation when multiple sources are present on the same band will be pointed out and the solutions found will be commented and discussed showing the results of real experiments carried out on a recording scenario. 1. Introduction The joint application of acoustic source localization (ASL) and scene video tracking is becoming an important support technology for hybrid audio-video platforms, as for example in video-conferencing or security systems in public places. For acoustic source localization positive array beamforming have been traditionally used with great success [4]. Nevertheless, positive array beamformers present certain inconveniences. These structures show non-neglectable side lobes which demand an important computational power to be reduced. The aperture of the main lobe is usually wide, and to narrow it either more sensors are needed or more power-demanding algorithms have to be used. On the other hand large number of sensors make signal acquisition and conditioning interfaces expensive, rendering the system expensive and difficult to install and operate. Negative beamformers, on the other side do not require a large number of sensors, and their extension to beam-like detection is straight forward. Nevertheless negative beamformers need to be adapted to broad-band signal enhancement and tracking. In a preliminary paper [2] the authors have shown how to track and separate individual sinusoidal sources in different bands, and to extend the results to broad-band (speech) signals. Through this paper the separation of sources on the same band is addressed. Section 2 reviews the basics of negative beamforming to deal with broad-band signals. In Section 3 the problem of multiple source separation is treated. Section 4 shows some practical study cases.

Most vocal and voice diseases cause changes in the voice. ENT clinicians use acoustic voice analysis to characterise pathological voices. The authors have focused their task in detection of impaired voices by means of neural network technology (ANN) and acoustic analysis. Former and actual works demonstrates that impaired voice detection can be carried out by means of supervised neural nets: MLP (multilayer perceptron). This paper is focussed in the task of detection of pathological voices by means of non-supervised neural nets (Kohonen self organising maps), comparing results with those obtained using supervised neural nets (MLPs). The aim of this paper is to study and compare two neural net based methods to be used for the detection of impaired voices: supervised (MLP ANNs) and non-supervised neural nets (Kohonen ANNs). Voice registers are parameterised by means of acoustic parameters.

We present a neural network application to the diagnosis of vocal and voice disorders, these disorders should be diagnosed in the early stage and normally cause changes in the voice signal. So we use acoustic parameters extracted from the voice as inputs for the neural network. In this paper, we focus our application on the distinction between pathologic and nonpathologic voices. The performance of the neural network is very good, 100% percent correct in the test. Furthermore, we have used neural network techniques to reduce the initial number of inputs (35), we conclude that only two acoustic parameters are needed for the classification between normal and pathological voices. The application can be a very useful diagnostic tool because it is noninvasive, makes possible to develop an automatic computer-based diagnosis system, is objective and can also be useful for evaluation of surgical, pharmacological and rehabilitation processes. Finally, we discuss the limitation of our work and possible future research.

Through the present paper, the evaluation results of a Speaker Independent Robust-to-Noise Isolated Word Recognizer are presented. The system, which is in part the results achieved by the project IVORY (ESPRIT project No. 20277) [6], is intended for working in highly non-stationary environments. The system comprises two main modules: the Noise Cancellator and the Speech Recognizer itself. System robustness is achieved in the noise cancellation module. This module incorporates an Adaptive Filter [7] [13], operating in the time domain, and a subsequent Spectral Subtraction step [2] operating in the frequency domain with the enhanced signal provided by the previous stage. Recognition results for different noise cancellation configurations and for several Parameter Extraction Front-Ends, including LPC [7], FFT Cepstrum [3] and PLP based methods [8] are presented. Keywords: robust speech recognition, non-stationary environments, recognition results. 1. INTRODUCTION Environmental noise is...

Most of vocal and voice diseases cause changes in the voice. These diseases have to be diagnosed and treated during an early stage. There is an increased risk for vocal and voice diseases due to the modern way of life. Acoustic voice analysis is an effective and non-invasive tool due to: a) Objective support of the diagnostics. b) Screening the vocal and voice diseases and especially their early detection. c) Objective determination of the impairment of the vocal function. d) Objective evaluation of the effect of the air pollution on the voice. e) Evaluation of surgical and pharmacological treatments. f) Evaluation of the rehabilitation. Many algorithms to calculate acoustic parameters have been developed and it is demonstrated that there is a great correlation between deviations of parameters and pathologies. The effectiveness and importance of the acoustic analysis of pathological voices has been proven by many experimental researches demonstrating that acoustic parameters of pathologic voices are deviated from the mean. The authors have focused their task in separation of pathologic/non pathologic voices, and evaluating the meaningful acoustic parameters by means of neural network technology and pruning methods.

The field of speech recognition in noisy environments is of vital importance for the success of certain applications in the domain of communications, automation, avionics, and other harsh situations where common recognizers fail to meet acceptable standards due to the negative influence of environmental noise collected during the recording of the speech trace. To overcome such difficulties, adaptive filtering has been traditionally used with acceptable success in noise cancellation using two-microphone schemes. Some problems found in such situations are due to the non-stationary behavior of speech and noise, although the sudden changes in the energy levels of both speech and noise are the cause of frequent misadjustments and un-locking in the adaptive algorithms, rendering the task of noise cancellation a difficult one. Through the present paper, a method based on the continuous tracking of the energy differences between speech and noise is presented, which may be used for the dynamic adaptation of the forgetting factor in the adaptive filters, producing an effective control of the sudden changes in the signal/noise ratio. The paper discusses the proposed methods and shows some results from practical conditions. These results show good stability and a special ability for word-boundary detection under highly non-stationary conditions, especially useful in isolated word speech recognition.

An augmented lattice is introduced for the analysis of production lines of robots with the analysis set up using binary Petri-nets. It is known that the lattice production line configuration of robots is about the most flexible for industrial manufacturing. This lattice configuration is slightly extended to further increase the flexibility. This augmented lattice is investigated using a single-place, multiple transition model for the robot information flow and the connection decision making. Using this model and binary Petri-net theory, the equations describing the system are obtained for a cascade of n augmented lattices. From the equations, a discussion is given on prevention of net deadlock, system tolerance, and production efficiency.