ResearchSpace

The effects of normalisation methods on speech emotion recognition

Show simple item record

dc.contributor.author Sefara, Tshephisho J
dc.date.accessioned 2020-03-17T13:08:55Z
dc.date.available 2020-03-17T13:08:55Z
dc.date.issued 2019-11
dc.identifier.citation Sefara, T.J. 2019. The effects of normalisation methods on speech emotion recognition. In: IEEE International Multidisciplinary Information Technology and Engineering Conference (IMITEC) 2019, Vanderbijlpark, South Africa, 21-22 November 2019 en_US
dc.identifier.isbn 978-1-7281-0040-1
dc.identifier.isbn 978-1-7281-0041-8
dc.identifier.uri https://ieeexplore.ieee.org/document/9015895
dc.identifier.uri DOI: 10.1109/IMITEC45504.2019.9015895
dc.identifier.uri http://hdl.handle.net/10204/11330
dc.description Presented at: IEEE International Multidisciplinary Information Technology and Engineering Conference (IMITEC) 2019, Vanderbijlpark, South Africa, 21-22 November 2019. This is the accepted version of the published item. en_US
dc.description.abstract Speech emotion recognition systems require features to be extracted from the speech signal. These features include Time, Frequency, and Cepstral-domain features. To normalise features, it is a challenging task to select an appropriate normalisation algorithm since the algorithm may impact classification accuracy. This paper presents the effects of different normalisation methods applied to speech features for speech emotion recognition. Speech features are extracted from the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset and normalised before training machine and deep learning algorithms such as Logistic Regression, Support Vector Machine, Multilayer Perceptron, Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). The CNNs and LSTMs obtained 72% for both accuracy and F1score outperforming standard machine learning algorithms. Feature normalisation improved both accuracy and F1score by more than 14% using CNN and LSTM. en_US
dc.language.iso en en_US
dc.relation.ispartofseries Workflow;23287
dc.subject Machine learning en_US
dc.subject Neural networks en_US
dc.subject Emotion recognition en_US
dc.subject Normalisation method en_US
dc.subject Speech emotions en_US
dc.title The effects of normalisation methods on speech emotion recognition en_US
dc.type Conference Presentation en_US
dc.identifier.apacitation Sefara, T. J. (2019). The effects of normalisation methods on speech emotion recognition. http://hdl.handle.net/10204/11330 en_ZA
dc.identifier.chicagocitation Sefara, Tshephisho J. "The effects of normalisation methods on speech emotion recognition." (2019): http://hdl.handle.net/10204/11330 en_ZA
dc.identifier.vancouvercitation Sefara TJ, The effects of normalisation methods on speech emotion recognition; 2019. http://hdl.handle.net/10204/11330 . en_ZA
dc.identifier.ris TY - Conference Presentation AU - Sefara, Tshephisho J AB - Speech emotion recognition systems require features to be extracted from the speech signal. These features include Time, Frequency, and Cepstral-domain features. To normalise features, it is a challenging task to select an appropriate normalisation algorithm since the algorithm may impact classification accuracy. This paper presents the effects of different normalisation methods applied to speech features for speech emotion recognition. Speech features are extracted from the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset and normalised before training machine and deep learning algorithms such as Logistic Regression, Support Vector Machine, Multilayer Perceptron, Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM). The CNNs and LSTMs obtained 72% for both accuracy and F1score outperforming standard machine learning algorithms. Feature normalisation improved both accuracy and F1score by more than 14% using CNN and LSTM. DA - 2019-11 DB - ResearchSpace DP - CSIR KW - Machine learning KW - Neural networks KW - Emotion recognition KW - Normalisation method KW - Speech emotions LK - https://researchspace.csir.co.za PY - 2019 SM - 978-1-7281-0040-1 SM - 978-1-7281-0041-8 T1 - The effects of normalisation methods on speech emotion recognition TI - The effects of normalisation methods on speech emotion recognition UR - http://hdl.handle.net/10204/11330 ER - en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record