The idea is to train a model to translate texts from a dataset of parallel texts

It's a simple command line application that you can build and launch directly.

See here the goals for this application and some ways of developpement

• Add a preprocessing step to deal with the data
• Create a method to evaluate the algorithm
• Improve the algorithm by taking in consideration the use of alignment meaning the position of words in both source and target sentences.

• Create a graphic interface to make the process easier.

• Create a web application with API and web framework and deploy it.

Build the new google Translate

• Gradle - Dependency Management
• Java 8 - Used to generate RSS Feeds

Feel free to contribute

• Alamart

This project is licensed under the MIT License - see the LICENSE.md file for details