I don't understand why do we use this method
float32 diag_norm(vector_t var, uint32 len) { float32 log_det; float32 p; uint32 i; log_det = 0; for (i = 0; i < len; i++) { if (var[i] <= 0) { E_FATAL("Sanity check var[i] > 0 failed: %e\n", var[i]); } log_det += log(var[i]); } p = len * log(2.0 * M_PI); return - 0.5 * (log_det + p); }
why isn't there any documentation about these methods ? !
You can learn Baum-Welch theory from the Rabiner's tutorial
A tutorial on hidden markov models and selected applications in speech recognition (1989) http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.131.2084
thanks .. and is this part of a full book , and how can I get the full book ?
Log in to post a comment.
I don't understand why do we use this method
why isn't there any documentation about these methods ? !
You can learn Baum-Welch theory from the Rabiner's tutorial
A tutorial on hidden markov models and selected applications in speech
recognition (1989)
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.131.2084
thanks .. and is this part of a full book , and how can I get the full book ?