I am trying to understand the analysis by synthesis in speech coding. In wikipedia it is mentioned for the CELP encoder:
The main principle behind CELP is called Analysis-by-Synthesis (AbS) and means that the encoding (analysis) is performed by perceptually optimizing the decoded (synthesis) signal in a closed loop.
What I cannot understand is, how is the encoding performed by optimizing the decoded code? Isn't the code decoded at the receiver?