The Emotion Probe: On the Universality of Cross-Linguistic and Cross-Gender Speech Emotion Recognition via Machine Learning.

Machine Learning (ML) algorithms within a human-computer framework are the leading force in speech emotion recognition (SER). However, few studies explore cross-corpora aspects of SER; this work aims to explore the feasibility and characteristics of a cross-linguistic, cross-gender SER. Three ML classifiers (SVM, Naïve Bayes and MLP) are applied to acoustic features, obtained through a procedure based on Kononenko's discretization and correlation-based feature selection. The system encompasses five emotions (disgust, fear, happiness, anger and sadness), using the Emofilm database, comprised of short clips of English movies and the respective Italian and Spanish dubbed versions, for a total of 1115 annotated utterances. The results see MLP as the most effective classifier, with accuracies higher than 90% for single-language approaches, while the cross-language classifier still yields accuracies higher than 80%. The results show cross-gender tasks to be more difficult than those involving two languages, suggesting greater differences between emotions expressed by male versus female subjects than between different languages. Four feature domains, namely, RASTA, F0, MFCC and spectral energy, are algorithmically assessed as the most effective, refining existing literature and approaches based on standard sets. To our knowledge, this is one of the first studies encompassing cross-gender and cross-linguistic assessments on SER.

Machine Learning-based Voice Assessment for the Detection of Positive and Recovered COVID-19 Patients.

Many virological tests have been implemented during the Coronavirus Disease 2019 (COVID-19) pandemic for diagnostic purposes, but they appear unsuitable for screening purposes. Furthermore, current screening strategies are not accurate enough to effectively curb the spread of the disease. Therefore, the present study was conducted within a controlled clinical environment to determine eventual detectable variations in the voice of COVID-19 patients, recovered and healthy subjects, and also to determine whether machine learning-based voice assessment (MLVA) can accurately discriminate between them, thus potentially serving as a more effective mass-screening tool. Three different subpopulations were consecutively recruited: positive COVID-19 patients, recovered COVID-19 patients and healthy individuals as controls. Positive patients were recruited within 10 days from nasal swab positivity. Recovery from COVID-19 was established clinically, virologically and radiologically. Healthy individuals reported no COVID-19 symptoms and yielded negative results at serological testing. All study participants provided three trials for multiple vocal tasks (sustained vowel phonation, speech, cough). All recordings were initially divided into three different binary classifications with a feature selection, ranking and cross-validated RBF-SVM pipeline. This brough a mean accuracy of 90.24%, a mean sensitivity of 91.15%, a mean specificity of 89.13% and a mean AUC of 0.94 across all tasks and all comparisons, and outlined the sustained vowel as the most effective vocal task for COVID discrimination. Moreover, a three-way classification was carried out on an external test set comprised of 30 subjects, 10 per class, with a mean accuracy of 80% and an accuracy of 100% for the detection of positive subjects. Within this assessment, recovered individuals proved to be the most difficult class to identify, and all the misclassified subjects were declared positive; this might be related to mid and short-term vocal traces of COVID-19, even after the clinical resolution of the infection. In conclusion, MLVA may accurately discriminate between positive COVID-19 patients, recovered COVID-19 patients and healthy individuals. Further studies should test MLVA among larger populations and asymptomatic positive COVID-19 patients to validate this novel screening technology and test its potential application as a potentially more effective surveillance strategy for COVID-19.

Voice analysis in adductor spasmodic dysphonia: Objective diagnosis and response to botulinum toxin.

INTRODUCTION: Adductor-type spasmodic dysphonia is a task-specific focal dystonia characterized by involuntary laryngeal muscle spasms. Due to the lack of quantitative instrumental tools, voice assessment in patients with adductor-type spasmodic dysphonia is mainly based on qualitative neurologic examination. We evaluated patients with cepstral analysis and specific machine-learning algorithms and compared the results with those collected in healthy subjects. In patients, we also used cepstral analysis and machine-learning algorithms to investigate the effect of botulinum neurotoxin type A. METHODS: We investigated 60 patients affected by adductor-type spasmodic dysphonia before botulinum neurotoxin type A therapy and 60 age and gender-matched healthy subjects. A subgroup of 35 patients was also evaluated after botulinum neurotoxin type A therapy. We recorded the sustained emission of a vowel and a sentence by means of a high-definition audio recorder. Voice samples underwent cepstral analysis as well as machine-learning algorithm classification techniques. RESULTS: Cepstral analysis was able to differentiate between healthy subjects and patients, but receiver operating characteristic curve analysis demonstrated that machine-learning algorithms achieved better results than cepstral analysis in differentiating healthy subjects and patients affected by adductor-type spasmodic dysphonia. Similar results were obtained when differentiating patients before and after botulinum neurotoxin type A therapy. Cepstral analysis and machine-learning measures correlated with the severity of voice impairment in patients before and after botulinum neurotoxin type A therapy. CONCLUSIONS: Cepstral analysis and machine-learning algorithms are new tools that offer meaningful support to clinicians in the diagnosis and treatment of adductor-type spasmodic dysphonia.

Towards the enhancement of body standing balance recovery by means of a wireless audio-biofeedback system.

Human maintain their body balance by sensorimotor controls mainly based on information gathered from vision, proprioception and vestibular systems. When there is a lack of information, caused by pathologies, diseases or aging, the subject may fall. In this context, we developed a system to augment information gathering, providing the subject with warning audio-feedback signals related to his/her equilibrium. The system comprises an inertial measurement unit (IMU), a data processing unit, a headphone audio device and a software application. The IMU is a low-weight, small-size wireless instrument that, body-back located between the L2 and L5 lumbar vertebrae, measures the subject's trunk kinematics. The application drives the data processing unit to feeding the headphone with electric signals related to the kinematic measures. Consequently, the user is audio-alerted, via headphone, of his/her own equilibrium, hearing a pleasant sound when in a stable equilibrium, or an increasing bothering sound when in an increasing unstable condition. Tests were conducted on a group of six older subjects (59y-61y, SD = 2.09y) and a group of four young subjects (21y-26y, SD = 2.88y) to underline difference in effectiveness of the system, if any, related to the age of the users. For each subject, standing balance tests were performed in normal or altered conditions, such as, open or closed eyes, and on a solid or foam surface. The system was evaluated in terms of usability, reliability, and effectiveness in improving the subject's balance in all conditions. As a result, the system successfully helped the subjects in reducing the body swaying within 10.65%-65.90%, differences depending on subjects' age and test conditions.

Associative memory design using support vector machines.

The relation existing between support vector machines (SVMs) and recurrent associative memories is investigated. The design of associative memories based on the generalized brain-state-in-a-box (GBSB) neural model is formulated as a set of independent classification tasks which can be efficiently solved by standard software packages for SVM learning. Some properties of the networks designed in this way are evidenced, like the fact that surprisingly they follow a generalized Hebb's law. The performance of the SVM approach is compared to existing methods with nonsymmetric connections, by some design examples.

Associative memory design for 256 gray-level images using a multilayer neural network.

A design procedure is presented for neural associative memories storing gray-scale images. It is an evolution of a previous work based on the decomposition of the image with 2L gray levels into L binary patterns, stored in L uncoupled neural networks. In this letter, an L-layer neural network is proposed with both intralayer and interlayer connections. The connections between different layers introduce interactions among all the neurons, increasing the recall performance with respect to the uncoupled case. In particular, the proposed network can store images with the commonly used number of 256 gray levels instead of 16, as in the previous approach.