Concealment of frame erasures for speech transmission and storage system and method

摘要

A decoder for packetized speech with differential quantization of line spectral frequencies and fixed-codebook gain conceals erased frames with interpolation of future and past frames by reconstruct future frame predicted parameters from presumed interpolations of erased frame parameters.