Skip to content

Latest commit

 

History

History
29 lines (20 loc) · 933 Bytes

README.md

File metadata and controls

29 lines (20 loc) · 933 Bytes

toipa

To ipa transformer based project. Experimentally turn a written language to ipa.

sk2ipa

Slovak language to ipa. This is an example of a language which is word based.

This is not really ipa as one letter is different which is easier for me to read. I think it was ť - θ, i prefer this to the ť - c.

Furthermore there are special placeholders. They are:

A - abbrevation. Should be expanded someday. F - foreign word, read as in the foreign language.

jp2ipa

Japanese to ipa. This is an example for a whole sententce being fed to the transformer language.

installation

pip install -r ./requirements.txt

Now you can train the models using the provided train.sh script. If your session is interrupted, you can resume later using the resume.sh script.

Once done, you can generate the dictionary for every word using the dict.sh script, or do inference for whole sentences using the eval.sh script.