Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Juraj Kacur has a total of 6 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Frequency, Time, Representation and Modeling Aspects for Major Speech and Audio Processing Applications

    by Juraj Kacur, Boris Puterka, Jarmila Pavlovicova, Milos Oravec

    There are many speech and audio processing applications and their number is growing. They may cover a wide range of tasks, each having different requirements on the processed speech or audio signals and, therefore, indirectly, on the audio sensors as well. This article reports on tests and evaluation of the effect of basic physical properties of speech and audio signals on the recognition accuracy of major speech/audio processing applications, i.e., speech recognition, speaker recognition, speech emotion recognition, and audio event recognition. A particular focus is on frequency ranges, time intervals, a precision of representation (quantization), and complexities of models suitable for each class of applications. Using domain-specific datasets, eligible feature extraction methods and complex neural network models, it was possible to test and evaluate the effect of basic speech and audio signal properties on the achieved accuracies for each group of applications. The tests confirmed that the basic parameters do affect the overall performance and, moreover, this effect is domain-dependent. Therefore, accurate knowledge of the extent of these effects can be valuable for system designers when selecting appropriate hardware, sensors, architecture, and software for a particular application, especially in the case of limited resources.


  • Open Access

    Algebraic Zero Error Training Method for Neural Networks Achieving Least Upper Bounds on Neurons and Layers

    by Juraj Kacur

    In the domain of artificial neural networks, it is important to know what their representation, classification and generalization capabilities are. There is also a need for time and resource-efficient training algorithms. Here, a new zero-error training method is derived for digital computers and single hidden layer networks. This method is the least upper bound on the number of hidden neurons as well. The bound states that if there are N input vectors expressed as rational numbers, a network having N − 1 neurons in the hidden layer and M neurons at the output represents a bounded function F: RD→RM for all input vectors. Such a network has massively shared weights calculated by 1 + M regular systems of linear equations. Compared to similar approaches, this new method achieves a theoretical least upper bound, is fast, robust, adapted to floating-point data, and uses few free parameters. This is documented by theoretical analyses and comparative tests. In theory, this method provides a new constructional proof of the least upper bound on the number of hidden neurons, extends the classes of supported activation functions, and relaxes conditions for mapping functions. Practically, it is a non-iterative zero-error training algorithm providing a minimum number of neurons and layers.

  • Open Access

    On the Speech Properties and Feature Extraction Methods in Speech Emotion Recognition

    by Juraj Kacur, Boris Puterka, Jarmila Pavlovicova, Milos Oravec

    Many speech emotion recognition systems have been designed using different features and classification methods. Still, there is a lack of knowledge and reasoning regarding the underlying speech characteristics and processing, i.e., how basic characteristics, methods, and settings affect the accuracy, to what extent, etc. This study is to extend physical perspective on speech emotion recognition by analyzing basic speech characteristics and modeling methods, e.g., time characteristics (segmentation, window types, and classification regions—lengths and overlaps), frequency ranges, frequency scales, processing of whole speech (spectrograms), vocal tract (filter banks, linear prediction coefficient (LPC) modeling), and excitation (inverse LPC filtering) signals, magnitude and phase manipulations, cepstral features, etc. In the evaluation phase the state-of-the-art classification method and rigorous statistical tests were applied, namely N-fold cross validation, paired t-test, rank, and Pearson correlations. The results revealed several settings in a 75% accuracy range (seven emotions). The most successful methods were based on vocal tract features using psychoacoustic filter banks covering the 0–8 kHz frequency range. Well scoring are also spectrograms carrying vocal tract and excitation information. It was found that even basic processing like pre-emphasis, segmentation, magnitude modifications, etc., can dramatically affect the results. Most findings are robust by exhibiting strong correlations across tested databases.


  • An Analysis of Eye-Tracking Features and Modelling Methods for Free-Viewed Standard Stimulus: Application for Schizophrenia Detection

    by Juraj Kacur, Jaroslav Polec, Eva Smolejova, Anton Heretik

    Currently psychiatry is a medical field lacking an automated diagnostic process. The presence of a mental disorder is established by observing its typical symptoms. Eye-movement specifics have already been established as a “endophenotype” for schizophrenia, but an automated diagnostic process of eye-movement analysis is still lacking. This article presents several novel approaches for the automatic detection of a schizophrenic disorder based on a free-view image test using a Rorschach inkblot and an eye tracker. Several features that enabled us to analyse the eye-tracker signal as a whole as well as its specific parts were tested. The variety of features spans global (heat maps, gaze plots), sequences of features (means, variances, and spectra), static (x and y signals as 2D images), dynamic (velocities), and model-based (limiting probabilities and transition matrices) categories. For each set of features, a proper modelling and classification method was designed (convolutional, recurrent, fully connected and combined neural networks; Hidden Markov models). By doing so, it was possible to find the importance of each feature and its physical representation using k-fold cross validation and a paired t-test. The dataset was sampled on 22 people with schizophrenia and 22 healthy individuals. The most successful approach was based on heat maps using all data and convolutional networks, reaching a 78.8% accuracy, which is a 10.5% improvement over the reference method. From all tested methods, there are two in an 85% accuracy range and over fifteen others in a 75% accuracy range at a 10% significance level.