METHOD AND DEVICES FOR CODING OR DECODING AN AUDIO SIGNAL OR BIT STREAM

摘要

In order to obtain a coded bit stream in a method for coding an audio signal, two time-discrete audio signal scan values are transformed in the frequency range with the purpose of obtaining spectral values. Said spectral values are coded using a code table with a limited number of code words having different lengths in order to obtain spectral values coded by code words, the length of a code word assigned to a spectral value being shorter if the probability of occurrence of the spectral value is higher. A raster is then determined for the coded bit stream, wherein the raster has equidistant grid points and the space between the grid points depends upon the code table(s) used. In order to obtain error-free Huffman encoding, priority code words representing given spectral values that are more significant than other spectral values from a psychoacoustic viewpoint are disposed in the raster in such a way that the beginning of the priority code word coincides with a grid point.