LPC INTERPOLATION BY APPROXIMATION OF THE SAMPLE AUTOCORRELATION FUNCTION

Conventionally, the energy of analysis frames is not taken into accoun t for linear prediction (LPC) interpolation. Incorporating the frame e nergy improves the subjective quality of interpolation, but increases the spectral distortion (SD). The main reason for this discrepancy is that the outliers are increased in low energy parts of segments with r apid changes in energy. The energy is most naturally combined with a n ormalized autocorrelation representation.