Adaptive block switching with deep neural networks

A method for obtaining at least one training block for training a computer implemented neural network system to predict at least one transform coefficient of an adaptive block length media signal, comprising:

• obtaining a set of transform blocks each comprising a number of transform coefficients representing frequency content of a media signal, the number of transform coefficients in each block being a first number or a second number, wherein the first number is greater than the second number, wherein a first subset of the set of transform blocks has the first number of transform coefficients and a second subset of the set of transform blocks has the second number of transform coefficients, and wherein the number of transform coefficients of a given block is related to a variance with respect to time of temporal characteristics of an audio signal associated with the given block,
• determining that a first block comprises the second number of transform coefficients,
• converting the first block into a converted block having the first number of transform coefficients,
• obtaining a target predicted block from the converted block,
• quantizing the converted block, and
• obtaining a training block from the quantized converted block.

A computer implemented neural network system for predicting transform coefficients representing frequency content of an adaptive block length media signal, said neural network system comprising:

• an adaptive block pre-processing unit configured to:
• receive a frame including two or more blocks, each block of the frame comprising a set of quantized transform coefficients representing a partial time segment of a media signal,
• receive block length information indicating a number of quantized transform coefficients for each block in said frame, the number of quantized transform coefficients being one of a first number or a second number, wherein said first number is greater than said second number, wherein a first subset of the two or more blocks has said first number of quantized transform coefficients and a second subset of the two or more blocks has said second number of quantized transform coefficients,
• determine that at least a first block has said second number of transform coefficients, and
• convert at least said first block into a converted block having said first number of quantized transform coefficients,
• a main neural network, wherein said main neural network is trained to predict at least one output variable given at least one conditioning variable based on conditioning information, said conditioning information comprising a representation of said converted block and a representation of block length information for said first block, and
• an output stage, configured to provide at least one predicted transform coefficient from said at least one output variable.