GMM implementation

Hi friends,

How do i implement gaussian mixture model for speaker recogntion. I have clustered the sample taken out mean and covariances also. I wanted to know how i can proceed further.I wanted to know wht parameters i can store in training phase and how do i test them to identify the person.

Thanks

Answer 1

[Tom Lane]

If you have the Statistics Toolbox, you may want to look at gmdistribution and its "cluster" method.