How to contribute to Nepali NLP

Tuesday. October 16, 2018 - 2 mins

Introduction:

This is a short tutorial to create English -> Nepali (and vice-versa) language translation. This is very important for all Nepali speaking people because this helps to create English -> Nepali language pair (and vice-versa) dataset, which eventually helps to move the Nepali NLP research ahead. This will help Nepali NLP researcher to focus on building stronger machine learning models specifically based on Nepali texts. Furthermore, these models can be used in text analysis, sentiment analysis, text classification, question-answering, speech recognition and much more.

I hope you are interested and excited as I am to contribute in this wonderful journey. The only thing that you need to do is translate the given English sentence into Nepali sentence.

Let’s dive into the technical details :-

You might want to create an account in Tatoeba . Tatoeba is a large database of sentences and translations. Its contents are created from the voluntary contributions of thousands of members all around the world.
Once you create an account, you might want to explore around a while. But, if you want to start contributing directly, then from the top menu, go to Contribute -> Translate Sentences. Click here.
In this page, set the settings as sentences in ENGLISH and not directly translated into NEPALI, then click on SHOW SENTENCES.
You will see a list of English sentences remained to be translated into Nepali language.
Once the sentences are displayed, click on of a particular sentence to add a translation.
Also click on to select nepali and press OK.
On another tab, open Easy Nepali Typing website which helps us to convert English texts into Nepali texts. For example: Ma nepal janchu -> म नेपाल जान्छु
In Easy Nepali Typing you need to type the English words with Nepali essence for the English sentence of your choice.
For example: if you selected I caught a cold today. from Tatoeba site. You need to type **malai aaja rugha laageko cha **.
As you type, it gets converted into मलाई आज रुघा लागेको छ ।
Once you get Nepali text, simply copy it and paste into previously opened Tatoeba site, select Language as Nepali and submit your translation.
Yayy!!!

I hope you enjoyed this tutorial. Thank you very much.

Keep contributing !!!

Many many thanks to Tatoeba and Easy Nepali Typing.