No pronunciation for word: (ALL!)

[Rita-ritally]

I have tried the whole process twice to make sure that there is no mistake in my environment. But it still wrong in
"bin/phonetisaurus-apply --model train/model.fst --word_list text.wlist"
what should i do......

[danijel3]

How did you create "model.fst"?

Oftentimes, it's important to use the same type of input for "text.wlist" as well as what was used in "model.fst". A common problem that happened to me was that the model was trained on all capital letters and when I used it on lowercase I would get this message.

[YangangCao]

I use lowercase letters to train model.fst, I also get bad results

$ ./phonetisaurus-apply  --model train/model.fst  --word_list oov.txt
WARNING:phonetisaurus-apply:2024-09-19 01:41:07:  No pronunciation for word: 'question'

$ ./phonetisaurus-apply  --model train/model.fst  --word_list oov.txt
WARNING:phonetisaurus-apply:2024-09-19 01:41:42:  No pronunciation for word: 'QUESTION'

[YangangCao]

phonetisaurus-g2pfst also get nothing

[YangangCao]

I find the solution, I didn't generate training lexicon in right way, the following way in README.md is good:

$ mkdir example
$ cd example
$ wget https://raw.githubusercontent.com/cmusphinx/cmudict/master/cmudict.dict
# Clean it up a bit and reformat:
$ cat cmudict.dict \
  | perl -pe 's/\([0-9]+\)//;
              s/\s+/ /g; s/^\s+//;
              s/\s+$//; @_ = split (/\s+/);
              $w = shift (@_);
              $_ = $w."\t".join (" ", @_)."\n";' \
  > cmudict.formatted.dict