The invention relates to a method for scalar quantization-based coding of the samples of a digital audio signal (S), the samples being coded over a pre-determined number of bits in order to obtain a binary frame of quantization indices (IPCM), the coding being carried out according to an amplitude compression law, where a pre-determined number of least significant bits are not taken into account in the binary frame of quantization indices. The coding method comprises the steps for storing (27) at least a part of the least significant bits that are not taken into account in the quantization index binary frame and for determination (28) of an enhancement bit stream (IEXT) comprising at least one bit thus stored.The invention also relates to an associated decoding method which comprises the steps for receiving (29) an enhancement bit stream (I′EXT) comprising one or more extension bits and for concatenation (30) of the extension bits behind the bits coming from the binary frame in order to obtain a decoded audio signal.The invention also relates to the coder and decoder implementing these methods.