Generalized Linear Discriminant Sequence for Speaker Verification
- Antoine Legrand
In speaker verification appplications, GMM models have an important place and have shown good perfomance. Actuallylinear discriminant methods using support vector machines (SVM) provide better results. We will focus on a linear disciminant system, the SVM-GLDS. Its uses statistics directly extracted from the speech features to define the recognition model without using Gaussian mixture models (GMMs). Weâll present and compare SVM-GLDS performance to SVM-GMM on NIST speaker evaluation tasks.