Channel compensation of speaker identification based on maximum a posteriori

To solve the effect of channel changes on the performance of speaker identification system, apply the method of maximum a posteriori to specific channel compensation. First, setup a standard channel, then, use the distribution of deviation of other channel and standard channel as experience knowledge, and obtain the deviation of current channel by a few adaptive data, use these deviations to adjust the characteristic to gain the goal of compensation. The results show that adaptive data in 1s, it has a discrimination of 89.1%, above the method of cepstrum average reduction by 4.2%, with the increasing of self-adaptive, the performance of this system can be improved.