Improving G.728´s hybrid window and excitation gain

The offset of gain filter is dealt unsuitably with a constant 32dB in G.728 standard. That is, the short-term stationary speech signal during the duration 20∼30ms is unsuitably supposed as stationary at whole time axis. It is necessary to design optimizing hybrid window for improving gain filter. In this work, the exact expression of excitation gain is conducted and an estimation method of SNR_av, used to optimize the parameters of hybrid window is presented. On the basis of above, this paper combines the Jayant adaptive quantization and G.728´s gain prediction to propose a scheme fit for designing gain codebook of LD-CELP, in which the exact gain value was adaptively predicted and quantized. In this method, for adaptive quantization gain of 3 bits and 4 bits, the SNR have been increased by 0.5 and 6 dB separately than that of G.728. If adopting 4 bits adaptive quantization and the shape codebook of 64 codevectors, the G.728´ complexity could be reduced by 20% while SNR had 1 dB improvement.