Methods of processing multi-channel audio signals, encoder and decoder

Abstract

The invention relates to a method for processing multi-channel audio signals, an encoder, and a decoder, and relates to the field of audio encoding and decoding techniques, to overcome the known technical limitations such that audio signals cannot be transmitted discontinuously in a multi-channel audio communication system. The encoder comprises a signal detection unit and a signal encoding unit. The signal encoding unit is configured so that: when the signal detection unit detects that the signal being downscaled to the Nth frame level includes a speech signal, the signal being downscaled to the Nth frame level is encoded; Alternatively, if the signal detection unit detects a signal that is downscaled to frame N excluding speech: it will encode the downscaled signal to frame N if the signal detection unit determines that the downscaled signal to frame N satisfies the predefined audio encoding conditions, or it will ignore the downscaled signal to frame N if the signal detection unit determines that the downscaled signal to frame N does not satisfy the predefined audio encoding conditions. According to this technical solution, because the encoding on the signal is discontinuous, the known technical problem that audio signals cannot be transmitted discontinuously is solved.