ubuntu@ubuntu-desktop:~$ perl prompts2wlist ./examininghtk/40/prompts.txt ./examininghtk/40/wlist
This created the wlist. Then I added SENT-END and SENT-START. Then I created global.ded. Copying the voxforge_lexicon to the folder. And now:
ubuntu@ubuntu-desktop:~$ cd ./examininghtk/40/
ubuntu@ubuntu-desktop:~/examininghtk/40$ HDMan -A -D -T 1 -m -w wlist -n monophones1 -i -l dlog dict voxforge_lexicon
Obviouly, the lexicon isn’t big enough. I should try a bigger lexicon VoxForgeDict. But this doesn’t solve my problem. There is something wrong with the punktuation, I think. I should remove the sentence marks. I did this with the gedit replace function.
And now (after moving prompts2wlist):
ubuntu@ubuntu-desktop:~/examininghtk/40$ perl prompts2wlist prompts.txt wlist
ubuntu@ubuntu-desktop:~/examininghtk/40$ HDMan -A -D -T 1 -m -w wlist -n monophones1 -i -l dlog dict VoxForgeDict
But 30 words are missing. I need to capitalize the words. Maybe I should try this script. This script is too simple, I need a different one. I did the conversion with OpenOffice.org. Result: “237 words processed, 19 missing” – that looks quite good.