In this project, I aim to create a hin-ben repository in Apertium that also includes the task of creating/expanding the transfer rules, creating the lexical selection rules, adding several thousand (in the range of 15,000 - 20,000) words in the hin-ben bidix. I also aim to test the translator on real texts to fine-tune the translator and presenting a finished translator with a WER of less than 25%, ready for publication, at the end of the project. I will also properly document the above mentioned deliverables.

Apertium is a free and open source project. The resources generated by this project will be generally available, not just for use as a translator. There has already been some work on this language pair, but it is unfinished. This project will generate a deliverable that will be a functional machine translator. This will not only make Apertium a functional Bengali morphological analyser/generator, which can be used for other language pairs, but will also give visibility to this fact by creating a functional translator based on Bengali.

Organization

Student

Gourab Chakraborty

Mentors

  • Hèctor Alòs i Font
close

2021