code prettify

Monday, 22 August 2016

Apertium: Free open source language translator

Came across a very cool "free" and "open source" tool to translate text from one language to another.




It's "Apertium" => "A free/open-source machine translation platform".

https://www.apertium.org

Excerpt from the site:
Apertium is a free/open-source machine translation platform, initially aimed at related-language pairs but expanded to deal with more divergent language pairs (such as English-Catalan). The platform provides a language-independent machine translation engine tools to manage the linguistic data necessary to build a machine translation system for a given language pair and linguistic data for a growing number of language pairs.
You can try the UI at https://www.apertium.org to translate sample text from one language to another.



Wiki and documentation related to it can be found at http://wiki.apertium.org/wiki/Main_Page

It also has a variety of tools for users / translators and developers: http://wiki.apertium.org/wiki/Tools

Apertium tools for users / translators

Apertium tools for developers


The installation steps are very easy. Just follow the one suitable for your OS at: http://wiki.apertium.org/wiki/Installation

Here is a video explaining the download, install process and command line usage of it for Ubuntu: https://www.youtube.com/watch?v=vy7rWy2u_m0

Below are the steps to install and use in a fresh new Ubuntu system:

1. Download stable-release sh script and run it to add Apertium package key to Ubuntu and to update the packages list.

$ wget https://apertium.projectjj.com/apt/install-release.sh -O - | sudo bash

2. To install all the core Apertium tools run:

$ sudo apt-get install apertium-all-dev

3. Download a language pair for your conversion use. You can search for your language pair here at http://wiki.apertium.org/wiki/List_of_language_pairs. For this example case, we will be downloading spanish-english language pair for conversion:

"apertium-en-es"  => English <-> Spanish

$ sudo apt-get install apertium-en-es

4. Now finally time to test it:

The command syntax is apertium <language-of-given-text>-<language-to-be-translated>

In our case, we will be translating from Spanish to English:

$ echo 'gracias' | apertium es-en

o/p: thank you

Now we will try to convert this spanish sentence "Tengo Un Mes Estudiando Español" which in english means "I've been learning Spanish for 1 month".

$ echo 'Tengo Un Mes Estudiando Español' | apertium es-en

o/p: Have A Month Studying Spanish

Same translation from google translate results in:

I have a month studying Spanish

Seems like Apertium translation result is quite similar to google translate result. This is great, coming from a free and open source tool. Hope you find use for it in your applications :)

No comments:

Post a Comment