VoxForge
>What are the acoustic models that can be used with Julius apart from
>acoustic models developed using HTK toolkit.
from the Julius website:
Julius adopts acoustic models in HTK ascii format, pronunciation dictionary in almost HTK format, and word 3-gram language models in ARPA standard format (forward 2-gram and reverse 3-gram trained from same corpus).
YOu might be interest in this post: Speech Recognition Model Converter
--- (Edited on 7/28/2010 4:11 pm [GMT-0400] by kmaclean) ---