Multichannel audio signal processing method, apparatus, and system

An encoder, comprising:

• a memory configured to store instructions;
• and
• a processor coupled to the memory, wherein the instructions cause the processor to be configured to:
• mix Nth-frame audio signals on two of a plurality of channels based on a first algorithm to obtain an Nth-frame downmixed signal;
• detect whether the Nth-frame downmixed signal comprises a speech signal, wherein N is a positive integer greater than zero;
• encode the Nth-frame downmixed signal when the Nth-frame downmixed signal comprises the speech signal;
• encode the Nth-frame downmixed signal when the Nth-frame downmixed signal satisfies a preset audio frame encoding condition and when detecting that the Nth-frame downmixed signal does not comprise the speech signal;
• and
• skip encoding the Nth-frame downmixed signal when the Nth-frame downmixed signal does not satisfy the preset audio frame encoding condition and when detecting that the Nth-frame downmixed signal does not comprise the speech signal.

A decoder, comprising:

• a memory configured to store instructions;
• and
• a processor coupled to the memory, wherein the instructions cause the processor to be configured to:
• receive a bitstream, wherein the bitstream comprises at least two frames, wherein the at least two frames comprise at least one first-type frame or at least one second-type frame, wherein the first-type frame comprises a downmixed signal, and wherein the second-type frame does not comprise the downmixed signal;
• and
• Nth-frame bitstream to obtain an Nth-frame downmixed signal when the Nth-frame bitstream is the first-type frame;
• and
• determine, according to a preset first rule, m-frame downmixed signals in at least one-frame downmixed signal preceding the Nth-frame downmixed signal, and obtain the Nth-frame downmixed signal according to the m-frame downmixed signals based on a first algorithm when the Nth-frame bitstream is the second-tyve frame, wherein m is a positive integer greater than zero, wherein N is a positive integer greater than one, and wherein the Nth-frame downmixed signal is received from an encoder after mixing Nth-frame audio signals on two of a plurality of channels based on a second algorithm.