nnmnkwii.autograd.mlpg¶

nnmnkwii.autograd.mlpg(means, variances, windows)[source]¶

Maximum Liklihood Paramter Generation (MLPG).

The parameters are almost same as nnmnkwii.paramgen.mlpg() expects. The differences are:

The function assumes means as torch.autograd.Variable instead of numpy.ndarray.
The fucntion assumes variances_frames as torch.FloatTensor instead of numpy.ndarray.

Parameters:	means (torch.autograd.Variable) – Means variances (torch.FloatTensor) – Variances windows (list) – A sequence of window specification