FILLING OF NON-CODED SUB-VECTORS IN TRANSFORM CODED AUDIO SIGNALS

摘要

A method of generating a virtual codebook (VC1) for filling non-coded residual sub-vectors of a transform coded audio signal below a predetermined frequency. The method including the steps of: quantizing (S1) coded residual sub-vectors, rejecting (S2) quantized residual sub-vectors that do not fulfill a predetermined sparseness criterion, and concatenating (S3) the remaining quantized residual sub-vectors to form the virtual codebook (VC1). The method further comprising combining (S4) pairs of coefficients of the first virtual codebook (VC1) to form a second virtual codebook (VC2) for filling non-coded residual sub-vectors of a transform coded audio signal above a predetermined frequency.