Montreal-Forced-Aligner
Montreal-Forced-Aligner copied to clipboard
[BUG]
Unfortantly I receive the following error, using both CMD and Colab, when I am trying to run the example from the documentation can be found here >> https://montreal-forced-aligner.readthedocs.io/en/latest/first_steps/example.html#alignment-example
" Error parsing line 0 of librispeech-lexicon.txt: Did not find any tabs, please ensure that your dictionary has tabs between words and their pronunciations. See output files at ./ljs_aligned "
And when I am look into the lexicon.txt file, I see that there aren't any tabs. only spaces. the lexicon has been download from the formal link of the documentation , https://montreal-forced-aligner.readthedocs.io/en/latest/first_steps/example.html#alignment-example
how this issue can be solved?
Thanks.
I have the same problem as you,and I tried to replace the english_us_arpa model and english_us_arpa dict, but it will arise the NoAlignmentsError。oh my god ,i am crazy to this tools。
I have the same problem.I using following code to modify librispeech-lexicon.txt and it works.
with open("librispeech-lexicon.txt",'r',encoding='utf-8') as f:
f_new = open("new_lexicon.txt",'w',encoding="utf-8")
lines = f.readlines()
for line in lines:
line_ = line.split("\t")
if len(line_)>1:
f_new.write(line)
else:
n = line.replace(" ","\t")
if len(n.split('\t')) !=1 and n.split('\t')[0]!="":
f_new.write(n)
f_new.close()
I have the same problem.I using following code to modify librispeech-lexicon.txt and it works.
with open("librispeech-lexicon.txt",'r',encoding='utf-8') as f: f_new = open("new_lexicon.txt",'w',encoding="utf-8") lines = f.readlines() for line in lines: line_ = line.split("\t") if len(line_)>1: f_new.write(line) else: n = line.replace(" ","\t") if len(n.split('\t')) !=1 and n.split('\t')[0]!="": f_new.write(n) f_new.close()
i change the dict,i think “lexcion.txt” can not suit the new version(2.0.6). you can dowanload the englis_us_arpa.dict