Biometric user authentication by voice is one of the most important functions of the information security. But, changes in the acoustic environment and communication channels create noise and various distortions in speech signals, whereby the recognition accuracy in such systems is considerably degraded. Therefore, removal of noise in speech signals is essential to improve the accuracy of speaker recognition. This paper proposes a method of adaptive noise reduction based on empirical wavelet transform, which is tested on speech signals with different noise levels (pp. 48-52).
- Daubechies , Lu J., Wu H.-T. Synchrosqueezed wavelet transforms: an empirical mode decomposition-like tool // Applied and Computational Harmonic Analysis, 2010, vol.30, no.2, pp.243–261.
- Wu H.-T., Flandrin P., Daubechies I. One or Two Frequencies? The Synchrosqueezing Answers // Advances in Adaptive Data Analysis, 2011, vol.3, no.1–2, pp.29–39.
- Gilles J. Empirical Wavelet Transform // IEEE Transactions on Signal Processing, 2013, vol.61, no.16, pp.3999–4010.
- Holambe R.S., Deshpande M.S. Noise robust speaker identification: using nonlinear modeling // Forensic Speaker Recognition, 2012, pp.153–182
- Imamverdiev Y.N., Suhostat L.V. Razrabotka robastnogo metoda izvlecheniya rechevyh priznakov na osnove ehmpiricheskogo vejvlet preobrazovaniya // Informacionnye tekhnologii, 2015, №1, c. 19–23.
- Schlotthauer G., Torres E., Rufiner H.L. A new algorithm for instantaneous F0 speech extraction based on Ensemble Empirical Mode Decomposition / Proc. of 17th European Signal Processing Conf., 2009, pp.2347–2351.
- Chhabra S., Bajaj R., Pachori R.B., Biswas R.N. Features based on Fourier-Bessel expansion for application of speaker identification system / Proc. Of Indian Conf. for Academic Research by Undergraduate Students, 2010, pp.1–3.
- Imamverdiyev Y.N., Sukhostat L.V. AZ-SRDAT - speech database for the Azerbaijan language // Problems of Information Technology, 2013, No 1, pp. 67-73.