Improving the Performance of Language Identification System Using Different GMM-Training Approaches

Wei Li; Dong Ju Kim; Kwang Seok Hong

doi:10.4028/www.scientific.net/AMR.121-122.496

Paper Titles

Study on Medical Micro-Image Mosaic with SIFT Features
p.476

A Network Intrusion Detection Method Inspired by Biological Immunology
p.482

A Negative Selection Algorithm Based on Hierarchical Clustering of Self Set
p.486

A Meme-Gene Co-Evolution Algorithm for the Image-Matching Question
p.490

Improving the Performance of Language Identification System Using Different GMM-Training Approaches
p.496

Bioinformatics Analysis of Laccase (Lac1) in Pycnoporus cinnabarinus
p.502

Dynamic Modeling of Tree Growth in a Nursery Greenhouse Using MATLAB and Simulink
p.507

Research of Evaluation Indicators about Electric Power Enterprise Marketing
p.512

Design and Development of Database System for Astronauts’ Dietary in Space Permanent Base
p.519

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 121-122Improving the Performance of Language...

Improving the Performance of Language Identification System Using Different GMM-Training Approaches

Abstract:

This paper proposed a feasible system for language identification (LID) and designed four different GMM-training approaches to improve the system performance by accuracy recognition rates. In our experiment, we used these model-training approaches to evaluation on the system performance, which utilizes Linear Prediction Cepstrum Coefficients (LPCC) and Gaussian Mixture Model (GMM), rely on a 10-language task. From all the results, we found an optimal approach for training GMM in LID system, which achieves high accuracy of 85.25%, and indicated that different GMM-training approaches have different performances for LID system, but an advisable training method that proposed in our paper can greatly improve the system performance.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 121-122)

Pages:

496-501

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.121-122.496

Citation:

Cite this paper

Online since:

June 2010

Authors:

Wei Li, Dong Ju Kim, Kwang Seok Hong

Keywords:

Acoustic Features, Language Identification, Statistical Model, Training Approaches

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Xiuhua Zeng, Jian Yang, Dan Xu. Approaches to Language Identification Using Gaussian Mixture Model and Linear Discriminant Analysis, IEEE CNF. Intelligent Information Technology Application Workshop, pp.1109-1112, (2008).

DOI: 10.1109/iita.workshops.2008.212

Google Scholar

[2] F. Allen, E. Ambikairajah and J. Epps, Language Identification Using Warping and the Shifted Delta Cepstrum, Proc. IEEE, Int Workshop on Multimedia Signal Processing (Shanghai, China), (2005).

DOI: 10.1109/mmsp.2005.248554

Google Scholar

[3] Wong, E. Sridharan, S. Comparison of Linear Prediction Cepstrum Coefficients and Mel-Frequency Cepstrum Coefficients for Language Identification, IEEE CNF. Intelligent Multimedia, Video and Speech Processing, pp.95-98, (2001).

DOI: 10.1109/isimp.2001.925340

Google Scholar

[4] Zbancioc, M. Costin, M. Using neural networks and LPCC to improve speech recognition, IEEE CNF. Computer Science Institute, Vol. 2, pp.445-448, (2003).

DOI: 10.1109/scs.2003.1227085

Google Scholar

[5] T. Toda, AW. Black and K Tokuda, Voice Conversion based on Maxium Likelihood Estimation of Spectral Parameter Trajectory, IEEE Trans. ASLP, vol. 15, No. 8, pp.2222-2235, Nov. (2007).

DOI: 10.1109/tasl.2007.907344

Google Scholar