the script is read the lexicon file, contains:
word spelling letter from word, example:
BETTER B E T T E R |
GOOD G O O D |
ANOTHER A N O T H E R |
when iterating that lexicon list (word and the spelling word) , It always error in my script.
the error is segmentation fault (core dump).
can somebody from the kenlm dev giving us a clue to solve it.
by the way, I build binary by clean the word. no character or something else, event number, already change to word. just word and space, then create binary from it.
not creating a trie file like in deepspeech, because, wav2letter / fairseq not ask me to do that.
Hello, I'm wondering..., how to solve segmentation error in inserting
spelling_idxs, word_idx, scoreto trie?the script is in here, line 156.
what i do is described in here.
the script is read the lexicon file, contains: word spelling letter from word, example:
when iterating that lexicon list (word and the spelling word) , It always error in my script. the error is segmentation fault (core dump).
can somebody from the kenlm dev giving us a clue to solve it.
by the way, I build binary by clean the word. no character or something else, event number, already change to word. just word and space, then create binary from it.
not creating a trie file like in deepspeech, because, wav2letter / fairseq not ask me to do that.
before it, really appreciate for the clue...