| 摘要 |
FIELD: information technology.SUBSTANCE: in an embedded signal encoding method, an embedded audio encoder receives an input signal to be encoded; the input signal is encoded through a first layer of the embedded audio encoder; a reconstructed first layer audio signal is obtained from the encoded input signal. Through a second layer of the embedded audio encoder, the reconstructed first layer audio signal is scaled with a plurality of gain values to obtain a plurality of scaled reconstructed audio signals, wherein this plurality of gain values depend on the reconstructed first layer audio signal and, also, each of said plurality of scaled reconstructed audio signals has an associated gain value; a plurality of error values are determined based on the input signal and each of said plurality of scaled reconstructed audio signals, and a gain value is selected from said plurality of gain values based on said plurality of error values. Through the embedded audio encoder, said gain value is transmitted or stored as part of the enhancement layer with respect to the encoded audio signal.EFFECT: improved quality of operation of CELP type encoders at low data transmission rates.13 cl, 7 dwg |
