Funding for two AI projects to prevent the ‘digital extinction’ of Irish – Siliconrepublic.com

Led by the Adapt research centre, the GaelTech and NRS projects are focused on machine translation technology.

Two technology projects have received more than €350,000 in Government funding to help prevent the “digital extinction” of the Irish language.

GaelTech and the National Relay Station (NRS) are looking to promote the digital preservation of Irish. The projects are led by Adapt, the Science Foundation Ireland research centre for AI-driven digital content technology, from Dublin City University (DCU).

The funding announced today (8 October) by Minister of State for Sport and the Gaeltacht Jack Chambers, TD, will see the two projects extended for the next two years.

One of more than 60 official minority languages recognised by the European Union, Irish has been classified as ‘definitely endangered’ by UNESCO. While most language technologies focus on widely used languages such as English and Chinese, Irish has limited digital support.

“New digital content tools allow minority languages to be supported in the same way through systems that can process and understand the language and hence help minority language speakers,” Adapt said in an announcement of the funding.

GaelTech

GaelTech is a project that was started in 2017 to develop natural language processing tools and resources for the Irish language. It aims to train the next generation of Irish language technologists to prevent the language from going extinct digitally.

“In the world of AI, investing in language technology is regarded as a step towards future-proofing a language against the risk of digital extinction,” lead researcher Dr Teresa Lynn said.

“As technology becomes more prevalent in our daily lives, we will need to ensure continued digital use of our minority languages, along with the necessary upskilling of those who can make this happen.”

Researchers will use the fresh funding to study the effect of automatic Irish multiword expression processing on features such as machine translation used in social media.

It will develop tools to understand the Irish language on social media and study its unique interaction with English on Twitter – research that can help users better express themselves in a language of their choice instead of defaulting to English.

National Relay Station

The other project, NRS, has developed a platform used by Irish language researchers to upload data to one centralised location for training machine translation engines. It was developed by Adapt researchers with funding from the EU’s European Language Resource Infrastructure  project.

Led by Adapt deputy director Prof Andy Way, NRS has collected data from various sources for its machine translation research in DCU. Way said that all machine translation systems today rely on high-quality parallel data – the original sentence and its translation – to ensure good quality translations.

The additional funding will allow the project to be hosted by DCU until the end of 2022 to support the development of translation technology at a national level. It will also look to improve the quality of Irish translations from the European Commission’s eTranslation system.

“Adapt at DCU has for a long time been engaged with Government departments to help provide a range of public documents in Irish for citizens who prefer to communicate through this language, rather than be forced to access information through English,” Way said.

He added that with Irish set to achieve full official working status the EU in 2022, this investment by the Government is “very timely indeed”.

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.