In general, reducing the data set is a good practice for development and debugging. If you want to build a quadratic discriminant model, you will need to invert these covariance matrices somehow. Then average the i-vectors across files to create an i-vector model for the speaker. ClassNames can have the following data types:. Creating Discriminant Analysis Model The model for discriminant analysis is: Each class Y generates data X using a multivariate normal distribution.
nest...