Unscented Transform for iVector-Based Noisy Speaker Recognition


Martinez, D., Burget, L., Stafylakis, T., Yun, L., Kenny, P., & Lleida, E. (2014, 4-9 May). Unscented transform for ivector-based noisy speaker recognition. Paper presented at the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP’14), Florence, Italy.


Recently, a new version of the iVector modelling has been proposed for noise robust speaker recognition, where the nonlinear function that relates clean and noisy cepstral coefficients is approximated by a first order vector Taylor series (VTS). In this paper, it is proposed to substitute the first order VTS by an unscented transform, where unlike VTS, the nonlinear function is not applied over the clean model parameters directly, but over a set of sampled points. The resulting points in the transformed space are then used to calculate the model parameters. For very low signal-to-noise ratio improvements in equal error rate of about 7% for a clean backend and of 14.50% for a multistyle backend are obtained.

Index Terms— Noise Robust Speaker Recognition, Unscented Transform, Vector Taylor Series, iVector

Read more from SRI