[BUG] Unknown error for "No files were aligned, this likely indicates serious problems with the aligner."

[hey0wing]

I had tried with the same file on older version at mac, but somehow it just couldn't work on the lastest version on win10.

Debugging checklist

[Y] Have you updated to latest MFA version? [Y] Have you tried rerunning the command with the --clean flag?

Describe the issue A clear and concise description of what the bug is.

For Reproducing your issue Please fill out the following:

1. Corpus structure

   • What language is the corpus in? Cantonese
   • How many files/speakers? 1
   • Are you using lab files or TextGrid files for input? no

2. Dictionary

   • Are you using a dictionary from MFA? If so, which one? no
   • If it's a custom dictionary, what is the phoneset? cantonese_pronunciation.txt

3. Acoustic model

   • If you're using an acoustic model, is it one download through MFA? If so, which one? no
   • If it's a model you've trained, what data was it trained on? no
   • It was found online from SPICE corpus https://dataverse.scholarsportal.info/dataset.xhtml?persistentId=doi:10.5683/SP2/MJOXP3

Log file Please attach the log file for the run that encountered an error (by default these will be stored in ~/Documents/MFA). validate_pretrained.log

Desktop (please complete the following information):

• OS: Windows
• Version Windows 10
• Any other details about the setup (Cloud, Docker, etc)

Additional context Add any other context about the problem here.

[WelkinYang]

perhaps due to too much oov words you can check the oov_list

[WelkinYang]

Now I found the reason for my alignment process with this error. It is because there is a clitic marker that is mistakenly included in the dictionary.txt. So the L.fst which is the input of compile-train-graphs-fsts.cc was not generated correctly. So maybe you need to check the dictionary. @mmcauliffe maybe you can add a checker program to make sure any specific makers in the config.yaml can not be a part of the dictionary.