VoxForge
Can anyone pls tell me how to do MAP using HTK?
Our aim is to convert a cluster of speaker independent models to Speaker Dependent models.
Thanks.
--- (Edited on 2/3/2011 9:42 pm [GMT-0600] by bharath5000) ---
>Can anyone pls tell me how to do MAP using HTK?
you might talk to someone at the falabrazil project.
The paper describing their project (An Open-Source Speech Recognizer for Brazilian Portuguese with a Windows Programming Interface) mentions that they used MLLR and MAP for acoustic model adaptation.
In addition, there is lots of code on their web site that might also provide some hints.
--- (Edited on 2/11/2011 11:25 pm [GMT-0500] by kmaclean) ---
I am trying to implement MAP adaptation in htk, but I can not find any example.
Is it enough to set the option "-u p" in HERest? Does MAP modify
directly the models or does it generate transforms like MLLR? Because I
can not find any transform.
Moreover.. to test the adaptation, can I use HVite normally on the model modified by MAP or should I use the transforms?
Thank you veryc much in advance!!
--- (Edited on 2/22/2013 6:05 am [GMT-0600] by Visitor) ---
I was happening the same problem.
I found the values of hmm files were changed when "-u mp" was set. "-u p" was not though.
Still I don't understand the difference between "mp" and "p"only.
--- (Edited on 6/19/2014 12:02 am [GMT-0500] by Visitor) ---