TCD Researchers Collaborate to Develop Tool Enabling Multilingual ‘Tweets’

Posted on: 14 June 2010

Researchers from Trinity College and Dublin City University launched an instant translation service for Twitter recently to coincide with the start of the World Cup in South Africa, allowing football fans share their thoughts and comments on the games in six European languages.  The system which instantly translates and streams tweets in Dutch, Portuguese, Spanish, Italian, French and German into English breaks down the language barrier for supporters in 22 of the 32 competing countries and encourages real-time lively debate on the highs and lows of the games.

Twitter is a social networking and micro blogging service that enables its users to send and read messages known as ‘tweets’.  Currently Twitter has over 100million users worldwide who increasingly like to discuss events as they happen.  More than half of all Twitter users are from non-English speaking countries and the development of the translation tool, known as Twanslator, is expected to encourage multilingual channels of communication between Twitter users culminating in a record number of tweets during the final stages of the games.

Available at www.myisle.org, Twanslator offers graphical tools to illustrate tweet frequencies and language breakdowns emerging from the Twitter World Cup stream.  The translated streams are then piped back onto Twitter under usernames that correspond to each of the supported languages.  This convenience allows Twitter users follow the translated streams through their Twitter client of choice.  Head of CNGL social media research, Dr Declan Dagger commented: “As events around the world unfold the social media ecosystem is first with the breaking news.  Through the CNGL, we are researching and developing next generation technologies that allow you filter out noise online ensuring you receive the content you want, when you want it and allows you decide how you want to consume it.”

Twanslator was a development between the Knowledge and Data Engineering Group (KDEG) at TCD’s School of Computer Science and Statistics, the machine translation and text analytics researchers at DCU and the group’s industrial partners at Microsoft.  The project, which is supported by Science Foundation Ireland, forms part of the Social Media initiatives of the Centre for Next Generation Localisation headed up by Deputy Director, Professor Vincent Wade of TCD’s School of Computer Science and Statistics.  The CNLG is focused on adapting digital content to language, culture and personal preference as localisation and personalisation is becoming a significant challenge within social media.