In this paper, a two-layer Multiple Kernel Learning (MKL) scheme for speaker-independent speech emotion recognition is presented. In the first layer, MKL is used for feature selection. The training samples are separated into n groups according to some rules. All groups are used for feature selection to obtain n sparse feature subsets. The intersection and the union of all feature subsets are the result of our feature selection methods. In the second layer, MKL is used again for speech emotion classification with the selected features. In order to evaluate the effectiveness of our proposed two-layer MKL scheme, we compare it with state-of-the-art results. It is shown that our scheme results in large gain in performance. Furthermore, another experiment is carried out to compare our feature selection method with other popular ones. And the result proves the effectiveness of our feature selection method.
Yun JIN
Southeast University,Jiangsu Normal University
Peng SONG
Southeast University
Wenming ZHENG
Southeast University
Li ZHAO
Southeast University
Minghai XIN
Southeast University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yun JIN, Peng SONG, Wenming ZHENG, Li ZHAO, Minghai XIN, "Speaker-Independent Speech Emotion Recognition Based on Two-Layer Multiple Kernel Learning" in IEICE TRANSACTIONS on Information,
vol. E96-D, no. 10, pp. 2286-2289, October 2013, doi: 10.1587/transinf.E96.D.2286.
Abstract: In this paper, a two-layer Multiple Kernel Learning (MKL) scheme for speaker-independent speech emotion recognition is presented. In the first layer, MKL is used for feature selection. The training samples are separated into n groups according to some rules. All groups are used for feature selection to obtain n sparse feature subsets. The intersection and the union of all feature subsets are the result of our feature selection methods. In the second layer, MKL is used again for speech emotion classification with the selected features. In order to evaluate the effectiveness of our proposed two-layer MKL scheme, we compare it with state-of-the-art results. It is shown that our scheme results in large gain in performance. Furthermore, another experiment is carried out to compare our feature selection method with other popular ones. And the result proves the effectiveness of our feature selection method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E96.D.2286/_p
Copy
@ARTICLE{e96-d_10_2286,
author={Yun JIN, Peng SONG, Wenming ZHENG, Li ZHAO, Minghai XIN, },
journal={IEICE TRANSACTIONS on Information},
title={Speaker-Independent Speech Emotion Recognition Based on Two-Layer Multiple Kernel Learning},
year={2013},
volume={E96-D},
number={10},
pages={2286-2289},
abstract={In this paper, a two-layer Multiple Kernel Learning (MKL) scheme for speaker-independent speech emotion recognition is presented. In the first layer, MKL is used for feature selection. The training samples are separated into n groups according to some rules. All groups are used for feature selection to obtain n sparse feature subsets. The intersection and the union of all feature subsets are the result of our feature selection methods. In the second layer, MKL is used again for speech emotion classification with the selected features. In order to evaluate the effectiveness of our proposed two-layer MKL scheme, we compare it with state-of-the-art results. It is shown that our scheme results in large gain in performance. Furthermore, another experiment is carried out to compare our feature selection method with other popular ones. And the result proves the effectiveness of our feature selection method.},
keywords={},
doi={10.1587/transinf.E96.D.2286},
ISSN={1745-1361},
month={October},}
Copy
TY - JOUR
TI - Speaker-Independent Speech Emotion Recognition Based on Two-Layer Multiple Kernel Learning
T2 - IEICE TRANSACTIONS on Information
SP - 2286
EP - 2289
AU - Yun JIN
AU - Peng SONG
AU - Wenming ZHENG
AU - Li ZHAO
AU - Minghai XIN
PY - 2013
DO - 10.1587/transinf.E96.D.2286
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2013
AB - In this paper, a two-layer Multiple Kernel Learning (MKL) scheme for speaker-independent speech emotion recognition is presented. In the first layer, MKL is used for feature selection. The training samples are separated into n groups according to some rules. All groups are used for feature selection to obtain n sparse feature subsets. The intersection and the union of all feature subsets are the result of our feature selection methods. In the second layer, MKL is used again for speech emotion classification with the selected features. In order to evaluate the effectiveness of our proposed two-layer MKL scheme, we compare it with state-of-the-art results. It is shown that our scheme results in large gain in performance. Furthermore, another experiment is carried out to compare our feature selection method with other popular ones. And the result proves the effectiveness of our feature selection method.
ER -