Calculate and classification analysis tool

Authors

Ing. Tomáš Kiska

Download

It is possible to download software here.

Publication to be cited

KISKA, T. Acoustic Analysis of Sentences Complicated for Articulation in Patients with Parkinson’ s Disease. In Proceedings of the 22nd Conference STUDENT EEICT 2016. Brno: Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních, 2016. s. 518-522. ISBN: 978-80-214-5350- 0.

Description

In the field pathological speech signal processing is necessary calculate speech features and also to use binary classification to evaluate the sufficiency of selected objective parametrization techniques to quantify and predict the health state of the analysed speaker (e.g. healthy speaker, disordered speaker). The researchers nowdays use multitude sophisticated classification techniques and metrices to evaluate classification models. Calculate and classification analysis tool (CaCAT) provides simply way to calculate speech features and sequential train and test of classification model.

Software CaCAT calculate 119 speech features describing especially articulation, but also phonation, prosody and speech fluency. It likewise provides classification techniques based on Classification Trees (RF) and metrics evaluating quality of classification proces: Classification Accuracy (ACC), sensitivity (SEN), specificity (SPE), trade-off between sensitivity and specificity (TSS), Mutual information (MI), Matthew’s Correlation Coefficient and Spearman’s rank sum correlation. For the purpose of feature selection CaCAT using mRMR (minimum Redundancy Maximum Relevance) method.

The CaCAT software is written in the MATLAB programming environment. The testing script VypoctiParametry script.m can be used to calculate features and preliminary individual statistical analysis. Testing script vypocetmRMR.m can be used to feature selection.

Projects

This work was supported by projects AZV 16-30805A, LO1401, COST IC1206 a FEKT-S-14-2335. The described research was performed in laboratories supported by the SIX project; the registration number CZ.1.05/2.1.00/03.0072, the operational program Research and Development for Innovation.

License

To negotiate the license terms of use of this software please contact the responsible person Ing. Lukas Novak at Technology Transfer Office, Brno University of Technology, Kounicova 966/67a, Veveří, 60200, Brno, Czech Republic, novak@ro.vutbr.cz.