kaldi::differentiable_transform Namespace Reference

Classes
class	AppendTransform
	This is a version of the transform class that consists of a number of other transforms, appended dimension-wise– e.g. More...
class	DifferentiableTransform
	This class is for speaker-dependent feature-space transformations – principally various varieties of fMLLR, including mean-only, diagonal and block-diagonal versions – which are intended for placement in the bottleneck of a neural net. More...
class	FmllrTransform
	Notes on the math behind differentiable fMLLR transform. More...
class	MinibatchInfoItf
class	NoOpTransform
	This is a version of the transform class that does nothing. More...
class	SequenceTransform
	This is a version of the transform class that does a sequence of other transforms, specified by other instances of the DifferentiableTransform interface. More...
class	SimpleMeanTransform
	This version of the transform class does a mean normalization: adding an offset to its input so that the difference (per speaker) of the transformed class means from the speaker-independent class means is minimized. More...
class	SpeakerStatsItf