![]() |
ITRW on Non-Linear Speech Processing
|
![]() |
Like almost all natural phenomena, speech is the result of many nonlinearly interacting processes; therefore any linear analysis has the potential risk of underestimating, or even missing, a great amount of information content. Recently the technique of Empirical Mode Decomposition (EMD) has been proposed as a new tool for the analysis for nonlinear and nonstationary data. We applied EMD analysis to decompose speech signal into intrinsic oscillatory modes. Besides, the LPC analysis of each mode provides an estimation of formants.
Bibliographic reference. Bouzid, Aïcha / Ellouze, Noureddine (2007): "EMD analysis of speech signal in voiced mode", In NOLISP-2007, 112-115.