Abstract
The noisy short utterance is polluted by noise and its corpus is not full, so the recognition rate significantly decreased. This paper proposed noise separation algorithm based on constrained Non-negative matrix factorization (CNMF), use it to separate pure speech from noisy speech. And then the speech frames are classified to high quality class and low quality class using differences detection and discrimination algorithm (DDADA) proposed in this paper. Combining features group with GMM-UBM two-stage classification model to make full use of limited information. Experiments show that the above algorithms improve speaker recognition rate of noisy short utterance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fatima, N., Zheng, T.F.: Short utterance speaker recognition. In: International Conference on Sustems and Informatics(ICSAI), pp. 1746–1750 (2012)
May, T., van de Par, S., Kohlrausch, A.: Noise-robust speaker recognition combining missing data techniques and universal background modeling. IEEE Transactions on Audio, Speech and Language Processing 20, 108–121 (2012)
Dehak, N., et al.: Front-end factor analysis for speaker verification. IEEE Transactions on Audio,Speech and Language Processing 19, 788–798 (2011)
Murty, K.S., Yegnanarayana, B.: Combining evidence from residual phase and MFCC features for speaker recognition. IEEE Signal Process 13, 52–55 (2006)
Chetouani, M., Faundez-Zanuy, M., Gas, B., Zarader, J.L.: Investigation on LP-residual representations for speaker identification. Pattern Recognition 42, 487–494 (2009)
Zheng, N., Ching, P.C., Lee, T.: Time frequency analysis of vocal source, signal for speaker recognition. In: Proc. ICSLP, pp. 2333–2336 (2004)
Chen, W.N., Zheng, N., Lee, T.: Discrimination power of vocal source and vocal tract related features for speaker segmentation. IEEE Transactions on Audio, Speech and Language Processing 15, 1884–1892 (2007)
Joder, C., Weninger, F., Eyben, F., Virette, D., Schuller, B.: Real-time speech separation by semi-supervised nonnegative matrix factorization. In: Proc. of Inter. Conf Latent Variable Analysis and Signal Separation (2012)
Joder, C., Schuller, B.: Exploring Nonnegative Matrix Factorization for Audio Classification: Application to Speaker Recognition. ITG-Fachbericht 236: Sprachkommunikation, Braunschweig (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chen, Y., Tang, ZM. (2013). The Speaker Recognition of Noisy Short Utterance. In: Sun, C., Fang, F., Zhou, ZH., Yang, W., Liu, ZY. (eds) Intelligence Science and Big Data Engineering. IScIDE 2013. Lecture Notes in Computer Science, vol 8261. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-42057-3_84
Download citation
DOI: https://doi.org/10.1007/978-3-642-42057-3_84
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-42056-6
Online ISBN: 978-3-642-42057-3
eBook Packages: Computer ScienceComputer Science (R0)