I see that evallm can generate .fprob files, and that interpolate can make new
ones from interpolating old ones. Both programs are in CMUCLTK. But after you
get .fprobs files as output, what can you do with them?
I think that perhaps that they can be used as input to create or modify
language model files.
Specifically, I might use probability streams to specialize a language model
for subsections of input. The sections are highly structured, and have
different vocabulary and likely grammar.
LAT
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I see that evallm can generate .fprob files, and that interpolate can make new
ones from interpolating old ones. Both programs are in CMUCLTK. But after you
get .fprobs files as output, what can you do with them?
I think that perhaps that they can be used as input to create or modify
language model files.
Specifically, I might use probability streams to specialize a language model
for subsections of input. The sections are highly structured, and have
different vocabulary and likely grammar.
LAT