In one embodiment, at least first and second channels in a frame of the audio signal are independently subdivided into blocks if the first and second channels are not correlated with each other. The first and second channels are corresponding subdivided into blocks such that the lengths of the blocks into which the second channel is subdivided correspond to the lengths of the blocks into which the first channel is subdivided if the first and second channels are correlated with each other. First information may be generated to indicate the subdivision of the channel into the blocks and the first information is generated commonly for first and second channels when the channels are correlated, and the first information is generated respectively for each of first and second channels when the channels are not correlated.