Wednesday, January 9, 2013

Open source Projects to do in Tamil computing

There are many projects to do in Open Source for Tamil Computing.

Listing here some of them.

1. Font conversion to unicode

There are lot of Tamil fonts in TSCII and ASCII format.

Example:
TAM, TAB, Vanavil, Bamini, indoword, softview, kabilan, kaniyan, shri TAM, Shri lipi, ilango, mayilai, anu, senthamizh etc. These are used in DTP centers.

There are tons of documents generated in these fonts.

To view them, we need to install these fonts locally.

These documents should be converted to Unicode so that anyone can view them without installing any special fonts.

NHM converter is a online service which does this.
http://software.nhm.in/services/converter

We need to create a FOSS application for this.

We can do this in python, php, ruby etc.

We can do this as desktop or web application.

TACE format should be considered.


2. Spellchecker

We can create a new spellchecker or extend the existing spellcheckers aspell or hunspell or project silpa.

Explore these:
www.silpa.in/

https://groups.google.com/forum/?fromgroups#!topic/freetamilcomputing/dEQgHESN9us

http://saranyaselvaraj.wordpress.com/2009/09/17/aspell-and-hunspell/


3. Grammer checker - santhi pizai thirutthi


4. Dictionary with Tamil meaning, english meaning, opposite, same meaning words

5. Number to string converter

example: 100 = nooru

6. OCR for Tamil

The following are in beginning stage.

http://gtamilocr.sourceforge.net

https://launchpad.net/tamilocr

test and extend them.


7. Tamil Corpus

A web application should be developed, showing a word and all the grammar tags.

Logged in users can select the relevant grammar tag for that word.

Thus, when many people contribute, a whole Tamil corpus will be generated.

8. Rule based auto complete for Tamil

9. Automatic machine translation

10. GUI and web based Tools to learn tamil for beginners

11. Text to Speech for Tamil

http://dhvani.sourceforge.net/

Ttest and enhance it

12. Project for Wiktionary

Wikionary is the wiki based dictionary for all languages.

Example:
http://ta.wiktionary.org

We can add voice files to wiktionary.

We need to create an web application, desktop and mobile client to display each word, asking the user to record the sound of the word.

Once recorded, the sound ogg file should be uploaded to commons.wikipedia.org and then it should be linked back to the same word in the wiktionary page.

Thus, any user can record and upload the audio words automatically.

13. In Tamil wikipedia, we need a javascript based on screen keyboard, so
that users can click and type easily.

Some of the projects are discussed in the following research paper collection.

http://ti2012.infitt.org/sites/default/files/Conference-book.part1.rar
http://ti2012.infitt.org/sites/default/files/Conference-book.part2.rar

Engineering college students who needs some base paper to their projects can use these papers and build applications on top of them.

Thanks to T.Shrinivasan, ILUGC.

1 comment:

  1. தமிழில் கலந்துள்ள ஆங்கில சொற்களை கண்டுபிடிப்பது எளிது ஆனால் வடமொழி சொற்களை கண்டுபிடிப்பது உண்மையில் கடினம்..சுத்த தமிழில் பேச வேண்டும் என்று தமிழ் ஆர்வலர்களே பேசும் அவலத்தை தடுக்க வேண்டும்..வடமொழி சொற்களை தட்டச்சு செய்யும் போது பிரித்துக்காட்டும் மென்பொருள் ஏதேனும் உள்ளதா..??

    ReplyDelete