Speech model parameter interpolation

A method of estimating speech model parameters from a digitized speech signal, the method comprising:

• dividing the digitized speech signal into two or more frequency band signals;
• determining a first excitation parameter at a first time sample, wherein determining the first excitation parameter comprises performing a nonlinear operation on at least two of the frequency band signals to produce at least first and second modified frequency band signals;
• determining a second excitation parameter at a second time sample, wherein determining the second excitation parameter comprises performing the nonlinear operation on the at least two of the frequency band signals to produce at least third and fourth modified frequency band signals;
• determining a first weight at the first time sample based on at least the first and second modified frequency band signals and corresponding voiced strengths, wherein determining the first weight comprises increasing the first weight when energy in at least one of the first and second modified frequency band signals near the first time sample increases;
• determining a second weight at the second time sample based on at least the third and fourth modified frequency band signals and corresponding voiced strengths, wherein determining the second weight comprises increasing the second weight when energy in at least one of the third and fourth modified frequency band signals near the second time sample increases;
• determining a third excitation parameter at a third time sample using the first and second excitation parameters and the first and second weights.