dc.contributor.author |
Sefara, Tshephisho J
|
|
dc.contributor.author |
Mokgonyane, TB
|
|
dc.date.accessioned |
2021-02-12T09:18:14Z |
|
dc.date.available |
2021-02-12T09:18:14Z |
|
dc.date.issued |
2020-11 |
|
dc.identifier.citation |
Sefara, T.J. & Mokgonyane, T. 2020. Emotional speaker recognition based on machine and deep learning. http://hdl.handle.net/10204/11753 . |
en_ZA |
dc.identifier.isbn |
978-1-7281-9521-6 |
|
dc.identifier.isbn |
978-1-7281-9520-9 |
|
dc.identifier.uri |
http://hdl.handle.net/10204/11753
|
|
dc.description.abstract |
Speaker recognition is a method which recognise a speaker from characteristics of a voice. Speaker recognition technologies have been widely used in many domains. Most speaker recognition systems have been trained on normal clean recordings, however the performance of these speaker recognition systems tends to degrade when recognising speech which has emotions. This paper presents an emotional speaker recognition system trained using machine and deep learning algorithms using time, frequency and spectral features on emotional speech database acquired from the Ryerson AudioVisual Database of Emotional Speech and Song (RAVDESS). We trained and compared the performance of five machine learning models (Logistic Regression, Support Vector Machine, Random Forest, XGBoost, and k-Nearest Neighbor), and three deep learning models (Long Short-Term Memory network, Multilayer Perceptron, and Convolutional Neural Network). After the evaluation of the models, the deep neural networks showed good performance compared to machine learning models by attaining the highest accuracy of 92% outperforming the state-of-the-art models in emotional speaker detection from speech signals. |
en_US |
dc.format |
Fulltext |
en_US |
dc.language.iso |
en |
en_US |
dc.relation.uri |
DOI: 10.1109/IMITEC50163.2020.9334138 |
en_US |
dc.relation.uri |
https://www.spu.ac.za/index.php/ieee-imitec-2020-programme/ |
en_US |
dc.relation.uri |
https://ieeexplore.ieee.org/xpl/conhome/9334048/proceeding |
en_US |
dc.source |
2nd International Multidisciplinary Information Technology and Engineering Conference (IMITEC), Sol Plaatje University, Kimberley, South Africa, 25 - 27 November 2020 |
en_US |
dc.subject |
Ryerson AudioVisual Database of Emotional Speech and Song |
en_US |
dc.subject |
RAVDESS |
en_US |
dc.subject |
Neural networks |
en_US |
dc.subject |
Machine learning |
en_US |
dc.subject |
Emotional recognition |
en_US |
dc.subject |
Speaker recognition |
en_US |
dc.title |
Emotional speaker recognition based on machine and deep learning |
en_US |
dc.type |
Conference Presentation |
en_US |
dc.description.pages |
8pp |
en_US |
dc.description.note |
Copyright: 2020 IEEE. This is the preprint version of the work. |
en_US |
dc.description.cluster |
Next Generation Enterprises & Institutions |
|
dc.description.impactarea |
Data Science |
en_US |
dc.identifier.apacitation |
Sefara, T. J., & Mokgonyane, T. (2020). Emotional speaker recognition based on machine and deep learning. http://hdl.handle.net/10204/11753 |
en_ZA |
dc.identifier.chicagocitation |
Sefara, Tshephisho J, and TB Mokgonyane. "Emotional speaker recognition based on machine and deep learning." <i>2nd International Multidisciplinary Information Technology and Engineering Conference (IMITEC), Sol Plaatje University, Kimberley, South Africa, 25 - 27 November 2020</i> (2020): http://hdl.handle.net/10204/11753 |
en_ZA |
dc.identifier.vancouvercitation |
Sefara TJ, Mokgonyane T, Emotional speaker recognition based on machine and deep learning; 2020. http://hdl.handle.net/10204/11753 . |
en_ZA |
dc.identifier.ris |
TY - Conference Presentation
AU - Sefara, Tshephisho J
AU - Mokgonyane, TB
AB - Speaker recognition is a method which recognise a speaker from characteristics of a voice. Speaker recognition technologies have been widely used in many domains. Most speaker recognition systems have been trained on normal clean recordings, however the performance of these speaker recognition systems tends to degrade when recognising speech which has emotions. This paper presents an emotional speaker recognition system trained using machine and deep learning algorithms using time, frequency and spectral features on emotional speech database acquired from the Ryerson AudioVisual Database of Emotional Speech and Song (RAVDESS). We trained and compared the performance of five machine learning models (Logistic Regression, Support Vector Machine, Random Forest, XGBoost, and k-Nearest Neighbor), and three deep learning models (Long Short-Term Memory network, Multilayer Perceptron, and Convolutional Neural Network). After the evaluation of the models, the deep neural networks showed good performance compared to machine learning models by attaining the highest accuracy of 92% outperforming the state-of-the-art models in emotional speaker detection from speech signals.
DA - 2020-11
DB - ResearchSpace
DP - CSIR
J1 - 2nd International Multidisciplinary Information Technology and Engineering Conference (IMITEC), Sol Plaatje University, Kimberley, South Africa, 25 - 27 November 2020
KW - Ryerson AudioVisual Database of Emotional Speech and Song
KW - RAVDESS
KW - Neural networks
KW - Machine learning
KW - Emotional recognition
KW - Speaker recognition
LK - https://researchspace.csir.co.za
PY - 2020
SM - 978-1-7281-9521-6
SM - 978-1-7281-9520-9
T1 - Emotional speaker recognition based on machine and deep learning
TI - Emotional speaker recognition based on machine and deep learning
UR - http://hdl.handle.net/10204/11753
ER - |
en_ZA |
dc.identifier.worklist |
24192 |
en_US |