Information Technology in an Improved Supervised Locally Linear Embedding for Recognizing Speech Emotion

Ri Sheng Huang

doi:10.4028/www.scientific.net/AMR.1014.375

Paper Titles

Applied-Information Technology in Robust Synchronization of Uncertain Fractional-Order Complex Networks
p.355

RETRACTED: Applied-Information Technology with a New Approach for Communication between AB SLC500 and Profibus-DP Network
p.359

Automatic Splicing Technology in Image Splicing Processing
p.367

Research on Electronic Measuring Technology with Development of Total Station Coordinate Traverse Adjustment Program
p.371

Information Technology in an Improved Supervised Locally Linear Embedding for Recognizing Speech Emotion
p.375

Analysis and Design of the Cloud System of Inner Mongolia National Cinema Line
p.379

Research of Digital TV Public Cultural Information Service System Based on DBS
p.383

Information Technology in Environment Digital Design Based on CAD
p.387

The Model Design of the Security of Electronic Records under Digital Office Environment
p.391

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 1014Information Technology in an Improved Supervised...

Information Technology in an Improved Supervised Locally Linear Embedding for Recognizing Speech Emotion

Abstract:

To improve effectively the performance on speech emotion recognition, it is needed to perform nonlinear dimensionality reduction for speech feature data lying on a nonlinear manifold embedded in high-dimensional acoustic space. This paper proposes an improved SLLE algorithm, which enhances the discriminating power of low-dimensional embedded data and possesses the optimal generalization ability. The proposed algorithm is used to conduct nonlinear dimensionality reduction for 48-dimensional speech emotional feature data including prosody so as to recognize three emotions including anger, joy and neutral. Experimental results on the natural speech emotional database demonstrate that the proposed algorithm obtains the highest accuracy of 90.97% with only less 9 embedded features, making 11.64% improvement over SLLE algorithm.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 1014)

Pages:

375-378

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.1014.375

Citation:

Cite this paper

Online since:

July 2014

Authors:

Ri Sheng Huang*

Keywords:

Manifold Learning, Nonlinear Dimensionality Reduction, Speech Emotion Recognition, Supervised Locally Linear Embedding

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] R. Picard. Affective computing. MIT Press, Cambridge, MA, pages 1-24, (1997).

Google Scholar

[2] J. Goddard, G. Schlotthauer and M. Torres, et al. Dimensionality reduction for visualization of normal and pathological speech data. Biomedical Signal Processing and Control, vol. 4, no. 3, pages 194-201, (2009).

DOI: 10.1016/j.bspc.2009.01.001

Google Scholar

[3] D. Yu. The application of manifold based visual speech units for visual speech recognition. [Ph.D. dissertation], Dublin City University, (2008).

Google Scholar

[4] D. De Ridder, O. Kouropteva and O. Okun, et al. Supervised locally linear embedding. Artificial Neural Networks and Neural Information Processing-ICANN/ICONIP-2003, Springer, Lecture Notes in Computer Science, vol. 2714, pages 333-341, (2003).

DOI: 10.1007/3-540-44989-2_40

Google Scholar

[5] D. Liang, J. Yang and Z. Zheng, et al. A facial expression recognition system based on supervised locally linear embedding. Pattern Recognition Letters, vol. 26, no. 15, pages 2374-2389, (2005).

DOI: 10.1016/j.patrec.2005.04.011

Google Scholar

[6] I. T. Jolliffe. Principal component analysis. New York: Springer, pages 150-165, (2002).

Google Scholar

[7] S. T. Roweis and L. K. Saul. Nonlinear dimensionality reduction by locally linear embedding. Science, vol. 290, no. 5500, pages 2323-2326, (2000).

DOI: 10.1126/science.290.5500.2323

Google Scholar

[8] J. B. Tenenbaum, V. Silva and J. C. Langford. A global geometric framework for nonlinear dimensionality reduction. Science, vol. 290, no. 5500, pages 2319-2323, 2000. 2319-2323.

DOI: 10.1126/science.290.5500.2319

Google Scholar