Just came across the iATROS speech decoder (GPLv3) - seems like it is similar to Julius in that it needs HTK for acoustic model creation. From their website:
iATROS is a new implementation of a previous speech recogniser that has been adapted to be used in both speech and handwritten text recognition. iATROS provides a modular structure that can be used to build different systems whose core is a Viterbi-like search on a Hidden Markov Model network. iATROS provides standard tools for off-line recognition and on-line speech recognition (based on ALSA modules).
--- (Edited on 9/8/2011 2:53 pm [GMT-0400] by kmaclean) ---