Encoding and decoding a portion of an audio signal

Abstract

Techniques are provided for encoding a portion of an audio signal. One procedure is performed by an audio encoder. The procedure comprises receiving an input vector comprising integer coefficients derived from said portion of the audio signal. The procedure comprises constructing an encoder tree by losslessly dividing the input vector into segments. Each segment is represented by a respective node in the encoder tree. Each node that is a head for another node comprises a sign information vector of the segments represented by the nodes that are leaves of other nodes. The segments represented by the leaf nodes have a sign set to a fixed sign value. The procedure comprises encoding said portion of the audio signal by enumerating the segments in a sequence of indices, with a unique index per segment.