System identification with adaptive lattice filters for speech data

Wenxia Shi, Vijay Parsa, Jagath Samarabandu

Abstract


The applications of adaptive lattice filter structures in modeling the response of hearing aids to speech signals were determined. Lattice based adaptive filter implementation include two approaches - the stochastic-gradient approach or gradient adaptive lattice (GAL), and the least-squares approach or least square lattice (LSL). Speech signals processed through modern hearing aids were analyzed using GAL filter and LSL filter. The performance of the algorithms were compared with least mean square (LMS), recursive least square (RLS), and affine projection algorithm (APA) in terms of computational complexity and modeling.

Keywords


Adaptive filtering; Computational complexity; Data reduction; Least squares approximations; Mathematical models; Speech communication; Systems analysis; Affine projection algorithms (APA); Least mean squares (LMS); Speech signals

Full Text:

PDF

Refbacks

  • There are currently no refbacks.