Cooperative pyramid vector quantizers for scalable audio coding

摘要

An example apparatus includes a memory configured to store the audio data; and one or more processors in communication with the memory, the one or more processors configured to: obtain, for each of a plurality of subbands of audio data, a respective energy scalar and a respective residual identifier; determine overall distortion levels for a plurality of candidate subband pulse allocations for performing pyramid vector dequantization (PVdQ) of the residual identifiers; select, from the plurality of subband pulse allocations and based on the overall distortion levels, a candidate subband pulse allocation; and perform, using the candidate subband pulse allocation, PVdQ on the residual identifiers to reconstruct a residual vector for each subband.