Delta vector taylor series environment compensation for speaker recognition

The performance of speaker recognition algorithms drops signi cantly when testing and training acoustic environments di er. This decrease is caused by the statistical mismatch between the statistics representing the speaker and the testing acoustic data. This paper reports our preliminary results on the application of a novel environmental compensation algorithm to the problem of speaker recognition and identi cation. This new technique, called the Delta Vector Taylor Series (DVTS) approach, improves performance at signal-to-noise ratios below 20dB. The algorithm imposes a model of how the environment modi es speaker statistics and uses ExpectationMaximization (EM) to solve a joint maximum likelihood formulation for the speaker recognition problem over both the speakers and the environment. We report experimental results on a subset of the TIMIT and NTIMIT database.