Linear prediction based on audio coding using an improved probability distribution estimator

摘要

Audio decoder based on linear prediction, comprising: a probability distribution estimator (102) configured to determine, for each plurality of spectral components, an estimate of probability distribution (28) from the linear prediction coefficient information contained in a data stream (22) in which the audio signal is encoded; an entropy decoding and decoding step (104) configured to decode and decrypt the entropy of a spectrum (26) composed of a plurality of spectral components from the data stream (22) using the probability distribution estimate (28) as determines each of the pluralities of spectral components; and a filter configured to shape the spectrum (26) according to a transfer function depending on a linear prediction synthesis filter defined by the linear prediction coefficient information, characterized in that the probability distribution estimator (102) is configured to determine a spectral fine structure (70) from the long-term prediction parameters contained in the data flow and to determine, for each of the pluralities of the spectral components, a probability distribution parameter so that the probability distribution parameters spectrally follow a function (72) that depends multiplicatively on the spectral fine structure (70), where, for each of the pluralities of spectral components, the probability distribution estimate (28) is a parameterizable function parameterized with the probability distribution parameter of the special component respective ctral.