جداسازي کور سيگنالهاي صحبت با الگوريتم بلوکي يکبعديDUET
محورهای موضوعی : مهندسی برق و کامپیوترسیدصادق فدائی 1 * , محمدحسین کهایی 2
1 - دانشگاه علم و صنعت ايران
2 - دانشگاه علم و صنعت ايران
کلید واژه: حوزه زمان-فرکانسالگوریتم DUETبرچسب تأخیر و تضعیفهیستوگرام,
چکیده مقاله :
جهت جداسازي سيگنالهاي صحبت به روش کور الگوريتم DUET استفاده ميشود که در آن با استفاده از ماسکنمودن حوزه زمان - فرکانس سيگنال هر منبع عمل جداسازي انجام ميشود. در اين الگوريتم به منظور محاسبه ماسکهاي زمان - فرکانس هيستوگرام دوبعدي پارامترهاي مخلوط تشکيل ميشود. اين روش زمان زيادي نياز دارد و به صورت غير بلادرنگ انجام ميشود. در اين مقاله، الگوريتمي پيشنهاد ميشود که عمل جداسازي را بصورت بلادرنگ انجام ميدهد. براي اين کار با الگوريتم DUET در حالت يکبعدي، پارامترهاي تأخير را تخمين ميزنيم. با اينکار حجم محاسبات به شدت کاهش يافته و اجراي بلادرنگ ممکن ميشود. نتايج شبيهسازيها نشان ميدهد که اين الگوريتم داراي دقت جداسازي قابل مقايسه با الگوريتم دوبعدي DUET بوده و به صورت بلادرنگ انجام ميشود.
To separate speech signals using blind techniques, the DUET algorithm is used in which each source signal is separated by masking the mixed signals in the Time-Frequency domain. To do so, a two dimensional Histogram of mixed parameters is generated which is computationally burden, and thus, can not be used in real-time. In this paper, we introduce a new algorithm in which the separation process can be carried out online. Also, simulation results show that this algorithm has a comparable precision with respect to the DUET algorithm.
[1] M. Baeck and U. Zolzer, "Real - time implementation of a source separation algorithm," in Proc. of the 6th Int. Conf. on Digital Audio Effects (DAFx’03), vol. 2, pp. 233-238, London, Sep. 2003.
[2] C. Jutten and J. Herault, "Blind separation of sources, part I: an adaptive algorithm based on neuromimetic architecture," in Signal Processing, vol. 24, pp. 1-10, Jul. 1991.
[3] A. Hyvarinen, J. Karhunen, and E. Oja, Independent Component Analysis, Wiley, New-York, 2001.
[4] J. Eriksson and V. Koivunen, "Identifiability, separability, and uniqueness of linear ICA models," in IEEE Signal Processing Letters, vol. 11, pp. 601-604, Jul. 2004.
[5] Z. M. Kamran, A. Rahim Leyman, and K. Abed–Meraim,"Techniques for blind source separation using higher-order statistics," in Proc. of the Tenth IEEE Workshop Statistical Signal and Array Processingpp. 334-338, Aug. 2000.
[6] A. Belouchrani, A. Meraim, J. F. Cardoso, and E. Moulines, "Ablind source separation technique using second - order statistics," IEEE Trans. Signal Processing, vol. 45, no. 2, pp. 434-444,Feb. 1997.
[7] D. T. Pham and J. F. Cardoso, "Blind separation of instantaneous mixtures of nonstationary sources," IEEE Trans. Signal Processing, vol. 49, no. 9, pp. 1837-1848, Sep. 2001.
[8] A. Belouchrani and M. G. Amin, "Blind source separation based on time - frequency signal representations," IEEE Trans. Signal Processing, vol. 46, no. 11, pp. 2888-2897, Nov. 1998.
[9] J. Ye, B. Xu, and F. Liu, "An improved cross - correlation based method of blind source separation in frequency domain," in Proc. IEEE Int. Conf. Neural Networks & Signal Processing, vol. 3,pp. 1370-1374, Dec. 2003.
[10] O. Yilmaz and S. Rickard, "Blind separation of speech mixtures via time-frequency masking," IEEE Trans. on Signal Processing, vol. 52, no. 7, pp. 1830-1847, Jul. 2004.
[11] A. Jourjine, S. Rickard, and O. Yilmaz, "Blind separation of disjoint orthogonal signals: demixing N sources from 2 mixtures," in Proc. IEEE Int. Conf. Acoustic, Speech, Signal Processing, vol. 5,pp. 2985-2988, Istanbul, Turkey, Jun. 2000.
[12] S. Rickard and O. Yilmaz, "On the approximate W – disjoint orthogonality of speech," in Proc. IEEE Int. Conf. Acoustic, Speech, Signal Processing, vol. 1, pp. 529-532, Orlando, US, May 2002.
[13] S. Rickard, R. Balan, and J. Rosca, "Real-time time-frequency based blind source separation," in Proc. Int. Workshop Independent Component Analysis Blind Source Separation, vol. 5, pp. 651-656,San Diego, US, Dec. 2001.
[14] http://alum.mit.edu/www/rickard/bss.html