Next: MMI training
Up: Training of a HMM
Previous: Training of a HMM
If we use the gradient techniques with ML, then the procedure would be
- Initialize the each HMM
with values generated randomly or using an initialization algorithm
like segmental K means .
- Take an observation sequence of a sentence and,
- Form the corresponding sentence model using the HMMs
of the speech units contained in the sentence.
- Calculate the forward and backward probabilities for
the sentence model, using the recursions 1.5 and
- Using the equation 1.21
calculate the likelihood of the observations in the
- Using the equations 1.26 and 1.29
calculate the gradients wrt all parameters in the
- Update parameters in each of the HMMs in the sentence
model using the eqn.1.19.
- Go to step (2), unless all the observation sequences are
- Repeat step(2) to (3) until a convergence criterion is
Fri May 10 20:35:10 MET DST 1996