Computation of segment Likelihood functions using PPT
In (14.2), the PDFs
are the segment PDFs
conditioned on knowing the segment sub-class (the
segment sub-classes are assumed to be fixed for each
segmentation ). These are obtained using the PPT (equation 2.2)
and a feature transformation specific to sub-class
on a segment of length base segments:
|
(14.4) |
where
is the feature used
for a segment of size and sub-class ,
is the PDF of that feature assuming sub-class ,
and
is the PDF of that feature
under the reference hypothesis.
This calculation has been described for a range of different
features (See chapters 4
through 11).
The MR-HMM requires a large amount of
front-end processing to calculate all the necessary features
and PDFs for all segments (seen in Figure 14.1 “Available segments")
assuming each possible sub-class.