Wikimedia collaborates with the Language Technologies Unit

A new collaboration between the Wikimedia Foundation and Bangor University has been announced that will enable both parties to share ideas and experiences from developing language technologies for multilingual situations. The collaboration will include integrating components of Bangor’s new TranslateWales.com software with the software that powers Wikipedia.

Bangor’s TranslateWales.com software is a new product for bilingual organizations, translation agencies and translators that’s been especially developed for optimized technological assistance in translating between Welsh and English. It includes a workflow manager, translation memories, glossaries as well as the more advanced components such as Cysill – the spelling and grammar checker - and Bangor’s own machine translation engines. TranslateWales.com enables professional translators to work more efficiently by providing machine translated text for correcting and final publication. This cuts down the work of human translators, whilst avoiding the pitfalls of pure machine translation.

David Chan, an expert on translation systems and a software engineer at Canolfan Bedwyr's Language Technologies Unit, said that the Wikimedia Foundation are developing a Content Translation tool for use by volunteer writers and editors and that Bangor’s machine translation engine will be part of the Wikimedia offering.    

David explained: “Although TranslateWales.com looks very different from Wikimedia’s multilingual environment, in reality much of our experience with it is useful in in designing the Wikimedia tool, especially the way machine translation integrates into the translators’ workflow. Translators are naturally worried that poor quality machine translation may be misused, but allowing the editor to view and edit the machine output overcomes that problem. ” David, who speaks Welsh, Cantonese, German as well as English added: “Welsh is really a medium-sized language in terms of world rankings: there are many languages which are much smaller in terms of both resources and number of speakers. It’s good that Welsh has indirectly been able to help some of these smaller languages.”

Alolita Sharma, the Director of Language Engineering at Wikimedia, said: “We welcome the opportunity to share Bangor University's expertise in machine translation and lesser-resourced language technology.”

Delyth Prys, the head of Canolfan Bedwyr's Language Technologies Unit added: “We’re really excited about this collaboration with Wikimedia and we hope to gain new insights and data that will help us improve our TranslateWales.com product and our machine translation engines. TranslateWales.com is beginning to take off in a big way in the public sector in Wales. We showcased it and our collaboration with Wikimedia at the recent Hacio’r Iaith event at Bangor University.”

Participants at the Hacio’r Iaith event, which included Wicipedia Cymraeg editors, professional developers and open source volunteers, as well as representatives from the public and private sectors, had the opportunity to hear more about the collaboration with Wikimedia as well as the opportunity to see and discuss the latest products, trends and ideas. Delyth Prys said “Hacio’r Iaith is the foremost Welsh language software and technology event in Wales and we very much enjoyed welcoming it to Bangor University. It was a great opportunity to discuss our own products and aspirations for the Welsh language and technology, but also to learn from others who are active in the field. We look forward to seeing our collaboration with the Wikimedia Foundation bearing fruit and are eager to collaborate with other organisations and companies.”

Publication date: 20 February 2014