Before using the service, please read the preliminary information containing a description of steps that enable access to the CLARIN-PL developer interface.
Dewulgaryzator is a service that allows you to replace vulgar expressions with their non-vulgar counterparts while maintaining the original character of the text. It is available for the Polish language and uses the DEPOTx tool developed by IPI PAN.
It is based on the t5-DEPOTxT5-base model.
This service can help maintain the appropriate level of discourse, and the deprofanated text can be useful wherever professionalism, understanding, and civil communication are key. With this service, the text can be made more appropriate and acceptable to a wider audience.
Dewulgaryzator can be run by using an LPMN query in the LPMN Client service:
No parameters.
Dewulgaryzator can be run in the Windows system with default values using the following LPMN query: ['any2txt','txt2txt'].
[['any2txt','txt2txt']] - input data in the form of a compressed directory (.zip)A text file.
A file containing the text with removed profanity.
In Colab: Dewulgaryzator - Replacement of profanity in the text
Cezary Klamra, Grzegorz Wojdyga, Sebastian Żurowski, Paulina Rosalska, Matylda Kozłowska & Maciej Ogrodniczuk (2022) "Devulgarization of Polish Texts Using Pre-trained Language Models", Computational Science – ICCS 2022. Lecture Notes in Computer Science, vol. 13351, Springer, Cham, 49--55.
(C) CLARIN-PL