INTERSPEECH 2011

This paper proposes a new approach to training the ivector space using a variant of PCA with the BaumWelch statistics for speaker verification. In eigenvoice the rank of variability space is bounded by the number of training speakers, so a variant of the probabilistic PCA approach is introduced for estimating the parameters. But this constraint doesn't exist in ivector model because the number of utterances is much bigger than the rank of total variability space. We adopt the EM algorithm for PCA with the statistics to train the total variability space, and the maximum likelihood criterion is used. After WCCN, the cosine similarity scoring is used for decision. These two total variability spaces will be fused at featurelevel and scorelevel. The experiments have been run on the NIST SRE 2008 data, and the results show that the performances in two total variability spaces are comparable. The performance can be improved obviously after feature fusion and score fusion.
Bibliographic reference. Lei, Zhenchun / Yang, Yingchun (2011): "Maximum likelihood ivector space using PCA for speaker verification", In INTERSPEECH2011, 27252728.