Multi-Information Model Fused Framework for Chinese Dialect Identification

Abstract:

This paper presents a model fused approach to Chinese dialect identification by combining multi-information including acoustic, phonotactic and prosodic feature. At first we analyze the way to translate language information into these features, and then propose a model fused framework for back-end classification. The experimental results show that the proposed method improves identification performance greatly and the prosodic features are more effective for shorter speech.

• Periodical: Applied Mechanics and Materials (Volumes 713-715)
• Pages: 2058-2062
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.713-715.2058
• Online since: January 2015
• Authors: Yu Guo Xia*, Ming Liang Gu
• Keywords: Chinese Dialect Identification, Information Fusing, Prosodic Feature

References

[1] M.A. Zissman. Comparison of four approaches to automatic language identification of telephone speech, IEEE Trans. Speech Audio Processing, 4(1): 31-44, (1996).

DOI: 10.1109/tsa.1996.481450

[2] F. Cummins, F. Gers and J. Schmidhuber, language identification from prosody without explicit features, Eurospeech , pp: 371-374, (1999).

DOI: 10.21437/eurospeech.1999-96

[3] Chang W. W. and Tsai W. H., Chinese dialect identification using segmental and prosodic features, Journal of Acoustic Society of America, 108(4): 1906-1913, (2000).

DOI: 10.1121/1.1289923

[4] B. Ma , Zhu D. L, and R. Tong , Chinese dialect identification using tone features based on pitch flux, IEEE Trans. ICASSP, vol. I: 1029-1032, (2006).

DOI: 10.1109/icassp.2006.1660199

[5] C.Y. Lin and H.C. Wang. Language identification using pitch contour information, IEEE Trans. ICASSP, vol. I: 601-604, (2005).

[6] P.A. Torres-Carrasquillo, D.A. Reynolds and R.J. Deller Jr. Language identification using Gaussian mixture model tokenization, IEEE Trans. ICASSP, vol. I: 757- 760, (2002).

DOI: 10.1109/icassp.2002.1005850

[7] B. P. Lim , H. Li and B. Ma. Using local and global phonotactic features in Chinese dialect identification, IEEE Trans. ICASSP, pp: 577-580, (2005).

DOI: 10.1109/icassp.2005.1415179

[8] R. Tong, B. Ma, D. Zhu, H. Li and E. S. Chng, Integrating acoustic, prosodic and phonotactic features for spoken language identification, IEEE Trans. ICASSP, pp.205-208, (2006).

DOI: 10.1109/icassp.2006.1659993

[9] Raymond W. M. Ng, T. Lee, C. C Leung, B Ma, H. Li: Spoken language recognition with prosodic features. IEEE Transactions on Audio, Speech, Language Processing, 21(9): 1841-1853, (2013).

DOI: 10.1109/tasl.2013.2260157

[10] Mingliang Gu, Zhaoyong Shen. Phonotatics based Chinese dialects identification. Journal of Chinese Information Processing, 20(5): 77-82, 2006. (In Chinese).