Embedded type coding, decoding method, encoder, decoder as well as system

Abstract

The embodiment of the invention provides an embedded coding method, an embedded decoding method, a coder, a decoder and a system, the decoding method comprises the following steps of: performing an algebraic code excitation linear prediction ACELP coding to a voice signal to obtain a code stream of an ACELP layer; performing transform domain coding to the voice signal to obtain the code stream of a transform domain layer. Because the ACELP coding excites the signal through gradual thinning, the coding signal with lower code rate is corresponding to the excitation signal comprising less algebraic pulses; along with the rise of the code rate, the manner increasing the algebraic pulses gradually thins the excitation signal ceaselessly, when the algebraic pulses are increased to a certain number, the thinning effect to the excitation signals is inconspicuous or weaker, therefore the embodiment of the invention uses the united coding manners of the ACELP and transform domain coding to obtain the embedded code stream, and uses the transform domain coding to compensate the ACELP coding, guaranteeing the quality of synthesis voice.

Description

technical field [0001] The invention relates to the field of coding and decoding, in particular to embedded coding, a decoding method, a coder, a decoder and a system. Background technique [0002] As the IP telephony technology based on packet switching gets more and more applications, the voice packet transmission technology develops rapidly. At present, an urgent problem to be solved in the voice packet transmission technology is how to reduce the impact of packet loss on the synthetic voice quality at the receiving end. During network transmission, some packets may be lost during transmission; some packets have exceeded the maximum allowable transmission delay when they arrive at the receiving end and are rejected by the receiving end; the receiving end simulates the lost packets according to the received packets , and then obtain the synthesized voice according to the received grouping and the simulated grouping, the quality of the synthesized voice obtained in this wa...

AI Technical Summary

Problems solved by technology

During network transmission, some packets may be lost during transmission; some packets have exceeded the maximum allowable transmission delay when they arrive at the receiving end and are rejected by the receiving end; the receiving end simulates the lost packets according to the received packets , and then the synthesized voice is obtained according to the received packets and the simulated packets. The quality of the synthesized voice obtained in this way will be greatly distorted compared with the original voice. When the lost packets reach a certain level, the synthesized voice cannot even be obtained.

Images