ADPCM quantizes the
difference between the sampled signal and a prediction. If the prediction is accurate, the difference between the real and predicted samples will have a lower variance than the real speech samples and will be accurately quantized with fewer bits than needed for the original.
At the decoder the quantized difference signal is added to the predicted signal to give the reconstructed speech signal. Performance is aided by using adaptive prediction and quantization, so that the predictor and difference quantizer adapt to the changing characteristics of the sampled speech.