Title :
Recursive estimation of wavelet based time varying AR models of speech
Author :
Reddy, K. Sreekanth ; Raja Kumar, R.V.
Abstract :
Estimation of time varying systems can be done by representing the whole system behaviour in terms of some fixed basis sequences. Available solutions use computationally expensive nonrecursive solutions. We derive an RLS solution for the wavelet based expansion of time varying systems. This method has been applied to speech signals and its performance is studied. A recursive lattice algorithm is derived for the above modeling. Also the performance of the method using both the Haar and Daubechies bases is studied.
Keywords :
Haar transforms; autoregressive processes; channel bank filters; least squares approximations; recursive estimation; sequences; speech processing; time-varying systems; wavelet transforms; Daubechies basis; Haar basis; Haar wavelet; RLS solution; autoregressive process; basis sequences; iterated filter bank; recursive estimation; recursive lattice algorithm; speech signals; time varying AR models; Equations; Filter bank; Lattices; Recursive estimation; Resonance light scattering; Signal processing algorithms; Speech analysis; Speech processing; System identification; Time varying systems;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on
Print_ISBN :
0-7803-7663-3
DOI :
10.1109/ICASSP.2003.1201680
