Every year small and medium enterprises spend hundreds of millions between them on translators and translation resources in an attempt to connect to international markets. Now a new research project will enable SMEs to cut dramatically the costs involved in translating their websites, through an exciting and innovative translation service called FALCON.
Website owners seeking high quality, low cost website translation and target market analytics are set to benefit from a localisation research project, FALCON, led by the ADAPT Centre for Digital Content Technology at Trinity College Dublin. The innovation of Dr David Lewis, Assistant Professor in Computer Science at Trinity, FALCON (Federated Active Linguistic data CuratiON) is a EU funded project that brings together a collection of European language technology developers and academics.
Using the semantic web concept, FALCON provides a fast and efficient mechanism for sharing translation memory and terminology data for specific domains that will be of interest to the many localisation companies from around the world.
“FALCON is a Software-as-a-Service (SaaS) translation tool chain that combines state-of-the-art website translation, translation management, computer aided translation and terminology management into a single offering”, explained Dr Lewis. “The solution is particularly suitable for SMEs and public bodies who may lack the expertise or language resource assets to tailor language technology to their needs. FALCON harnesses the synergy between the human language worker, data-driven language technologies and translation and terminology data resources to overcome the challenges of content volume, velocity and variety. This helps the language worker keep up with the surging demand for high-quality translation.”
Many companies competing globally are faced with localisation challenges that are expensive and risky to address. With demand for translation services growing by approximately 8% annually, those companies who can apply state-of-the-art machine translation technology efficiently, where computers learn from huge databases of already translated text to make ever better guesses about how to render whole chunks from one language into another, will pay less for the same amount of translation work. The FALCON project provides the foundation for effective statistical machine translation that is set to become the primary mode of translation in the future.
FALCON operates by combining the power of open data on the web with data-driven language technologies to construct the ‘Localisation Web’, allowing for rich connections of data between web systems across the globe. This network of terms and translations are inter-linked to each other, and used to source and target documents via URLs. FALCON integrates the resulting web of linked localisation and language data into its novel localisation tool chains using existing data model, query and access control standards. This tool chain integrates open source automatic text extraction and machine translation technology and public language resources. Iterative quality improvement in this language technology is delivered by using linked data to actively manage the curation and reuse language resources within customer projects.
“The advances seen in machine translation as a result of the FALCON project will contribute towards an expansion in global trade and increased competitiveness for SMEs worldwide,” commented Dr Lewis.
The FALCON project, funded by the EU, brings together the expertise of the Science Foundation Ireland funded ADAPT Centre, with the technical and commercial knowhow of three international SMEs from the localisation tool industry – Easyling, Interverbum and XTM International.
Share this article:
A digital twin is a virtual replica of a physical space. Such a replica is created with technologies like multiple-sensor #IoT, advanced #DataAnalytics & #AI. Register to catch @niallmurr share insights & research about it at the #AI Summer School https://www.eventbrite.ie/e/john-mccarthy-artificial-intelligence-summer-school-2022-ai-for-tourism-tickets-392472003327