scholarly journals GMM-based speaker age and gender classification in Czech and Slovak

2017 ◽  
Vol 68 (1) ◽  
pp. 3-12 ◽  
Author(s):  
Jiří Přibil ◽  
Anna Přibilová ◽  
Jindřich Matoušek
Keyword(s):  
Classification Accuracy ◽  
Gaussian Mixture Models ◽  
Gaussian Mixture ◽  
Age And Gender ◽  
Subjective Evaluations ◽  
Listening Tests ◽  
Different Types ◽  
Speech Features ◽  
And Gender

Abstract The paper describes an experiment with using the Gaussian mixture models (GMM) for automatic classification of the speaker age and gender. It analyses and compares the influence of different number of mixtures and different types of speech features used for GMM gender/age classification. Dependence of the computational complexity on the number of used mixtures is also analysed. Finally, the GMM classification accuracy is compared with the output of the conventional listening tests. The results of these objective and subjective evaluations are in correspondence.

Two-Level Classification in Determining the Age and Gender Group of a Speaker

2021 ◽  
Vol 18 (5) ◽  
Author(s):  
Ergün Yücesoy
Keyword(s):  
Classification Accuracy ◽  
Age Groups ◽  
Gaussian Mixture ◽  
Support Vector ◽  
Optimum Number ◽  
Mel Frequency Cepstral Coefficients ◽  
Age And Gender ◽  
Number Of Components ◽  
And Gender

In this study, the classification of the speakers according to age and gender was discussed. Age and gender classes were first examined separately, and then by combining these classes a classification with a total of 7 classes was made. Speech signals represented by Mel-Frequency Cepstral Coefficients (MFCC) and delta parameters were converted into Gaussian Mixture Model (GMM) mean supervectors and classified with a Support Vector Machine (SVM). While the GMM mean supervectors were formed according to the Maximum-a-posteriori (MAP) adaptive GMM-Universal Background Model (UBM) configuration, the number of components was changed from 16 to 512, and the optimum number of components was decided. Gender classification accuracy of the system developed using aGender dataset was measured as 99.02% for two classes and 92.58% for three classes and age group classification accuracy was measured as 67.03% for female and 63.79% for male. In the classification of age and gender classes together in one step, an accuracy of 61.46% was obtained. In the study, a two-level approach was proposed for classifying age and gender classes together. According to this approach, the speakers were first divided into three classes as child, male and female, then males and females were classified according to their age groups and thus a 7-class classification was realized. This two-level approach was increased the accuracy of the classification in all other cases except when 32-component GMMs were used. While the highest improvement of 2.45% was achieved with 64 component GMMs, an improvement of 0.79 was achieved with 256 component GMMs.

Poster 386 Evaluating Personal Factors of Age and Gender within the International Classification of Functioning, Disability, and Health (ICF)

PM&R ◽  
2016 ◽  
Vol 8 (9) ◽  
pp. S287
Author(s):  
Elisabeth Fehrmann ◽  
Simone Kotulla ◽  
Thomas Kienbacher ◽  
Patrick Mair ◽  
Josef Kollmitzer ◽  
...  
Keyword(s):  
International Classification ◽  
International Classification Of Functioning ◽  
Personal Factors ◽  
Disability And Health ◽  
Age And Gender ◽  
And Gender

Classification of a known sequence of motions and postures from accelerometry data using adapted Gaussian mixture models

2006 ◽  
Vol 27 (10) ◽  
pp. 935-951 ◽  
Author(s):  
Felicity R Allen ◽  
Eliathamby Ambikairajah ◽  
Nigel H Lovell ◽  
Branko G Celler
Keyword(s):  
Mixture Models ◽  
Gaussian Mixture Models ◽  
Gaussian Mixture

Classification and Treatment of Supramalleolar Deformities

2011 ◽  
Vol 32 (11) ◽  
pp. 1023-1031 ◽  
Author(s):  
Markus Knupp ◽  
Sjoerd A.S. Stufkens ◽  
Lilianna Bolliger ◽  
Alexej Barg ◽  
Beat Hintermann
Keyword(s):  
Ankle Joint ◽  
Operative Procedure ◽  
Level Of Evidence ◽  
Age And Gender ◽  
Additional Surgery ◽  
Different Types ◽  
Prospective Comparative Study ◽  
And Gender ◽  
Classification And Treatment

Background: Supramalleolar osteotomies are increasingly popular for addressing asymmetric arthritis of the ankle joint. Still, recommendations for the indication and the use of additional procedures remain arbitrary. We preoperatively grouped different types of asymmetric arthritis into several classes and assessed the usefulness of an algorithm based on these classifications for determining the choice of supramalleolar operative procedure and the risk factors for treatment failure. Methods: Ninety-two patients (94 ankles) were followed prospectively and assessed clinically and radiographically 43 months after a supramalleolar osteotomy for asymmetric arthritis of the ankle joint. Results: Significant improvement of the clinical scores was found. Postoperative reduction of radiological signs of arthritis was observed in mid-stage arthritis. Age and gender did not affect the outcome. Ten ankles failed to respond to the treatment and were converted to total ankle replacements or fused. Conclusions: Supramalleolar osteotomies can be effective for the treatment of early and midstage asymmetric arthritis of the ankle joint. However, certain subgroups have a tendency towards a worse outcome and may require additional surgery. Therefore preoperative distinction of different subgroups is helpful for determination of additional procedures. Level of Evidence: II, Prospective Comparative Study

scholarly journals New Families in our Solar Neighborhood: Applying Gaussian Mixture Models for Objective Classification of Structures in the Milky Way and in Simulations

2021 ◽  
Vol 921 (2) ◽  
pp. 106
Author(s):  
Farnik Nikakhtar ◽  
Robyn E. Sanderson ◽  
Andrew Wetzel ◽  
Sarah Loebman ◽  
Sanjib Sharma ◽  
...  
Keyword(s):  
Mixture Models ◽  
Milky Way ◽  
Gaussian Mixture Models ◽  
Gaussian Mixture ◽  
Solar Neighborhood ◽  
Objective Classification

scholarly journals Electrocardiogram (ECG) based stress recognition integrated with different classification of age and gender

2019 ◽  
Vol 15 (1) ◽  
pp. 199
Author(s):  
N. S. Nor Shahrudin ◽  
K. A. Sidek ◽  
A. Z. Jusoh
Keyword(s):  
Data Acquisition ◽  
Stress Level ◽  
Ecg Signal ◽  
Alternative Mechanism ◽  
Age And Gender ◽  
Stress States ◽  
And Gender ◽  
Electrocardiogram Ecg ◽  
Ecg Data

<p class="Abstract"><em><span>Good mental health is important in our daily life. A person commonly finds stress as a barrier to enhance an individual’s performance. Be reminded that not all people have the same level of stress because different people have dissimilar problems in their life. In addition, different level of age and gender will affect unequal amount of stress. Electrocardiogram (ECG) signal is an electrical indicator of the heart that can detect changes of human response which relates to our emotions and reactions. Thus, this research proposed a non-intrusive detector to identify stress level for both gender and different classification of age using the ECG. A total of 30 healthy subjects were involved during the data acquisition stage. Data acquisition which initialize ECG data were divided into two conditions; which are normal and stress states. ECG data for normal state only need the participant to breath in and out normally. In other hand, the participants also need to undergo Stroop Colour word test as a stress inducer to represent ECG in stress state. Then, Sgolay filter was selected in the pre-processing stage to remove artifacts in the signal. The process was followed by feature extraction of the ECG signal and finally classified using RR Interval (RRI), different amplitudes of R peaks and Cardioid graph were used to evaluate the performance of the proposed technique. As a result, Class 5 (age range between 50-59 years old) marks the highest changes of stress level rather than other classes, while women are more affected by stress rather than men by showing tremendous percentage changes between normal and stress level over the proposed classifiers. The result proves that ECG signals can be used as an alternative mechanism to recognize stress more efficiently with the integration of gender and age variabilities.</span></em></p>

Sign in / Sign up

Export Citation Format

Share Document