Speaker adaptations in sparse training data for improved speaker verification

The over-training problem in speaker verification occurs when modelling a speaker with sparse training data. The authors propose to solve this problem by employing effective speaker adaptations, using a hybrid version of the maximum a posteriori and maximum likelihood linear regression methods. Experimental results show that the speaker verification system, using the proposed hybrid adaptation scheme, outperforms systems based on speaker models without adaptation by a factor of up to 5.