A method for an audio processing system, applied to an electronic device acquiring a first audio processing system, the first audio processing system including a first encoding network and a first decoding network, generating a second encoding network corresponding to the first encoding network based on a configuration request for the first encoding network, encoding a first audio sample based on the second encoding network to obtain an audio bit-stream sample of the first audio sample, decoding the audio bit-stream sample based on the first decoding network to obtain a reconstructed first audio sample of the first audio sample, and updating parameters of the second encoding network based on the reconstructed first audio sample to obtain a trained second encoding network.