Language Technologies & Multilingualism

How can we bring online content to everyone, no matter where they are from and what language they speak? How can local, regional and national public administrations across the EU make forms available and interact with citizens in multiple languages? How can small and medium-sized enterprises (SMEs) quickly access markets beyond their national ones and fully benefit from the Single Market?

The obvious answer is through language technologies: we need to teach computers how to understand and process written and spoken human language, build fit for purpose applications and deploy them widely.

Language technologies go far beyond machine translation. They offer applications for text analysis, such as named-entity recognition and anonymisation, dialogue systems, search engines, automatic text summarisation, speech-to-text and more. Language technologies can be developed and customized for any specific scenario where human language is processed.

Addressing the language barriers in the digital sphere requires European action on various levels:

• the Horizon Europe Programme fosters research and innovation through cross-sectorial support of language technologies;
• the Digital Europe Programme encourages European public and private sectors to deploy language technologies.

The European Commission has developed a range of basic language tools. These are created using the newest AI technologies and have been trained on the large amounts of data available both in-house and gathered through an EU-wide language resource collection effort.

Moreover, European language technology service providers offer excellent customized solutions for specific areas and domains.

If we Europeans join forces and consolidate our efforts, we put ourselves in a position to deploy excellent solutions, both publicly and privately based, thereby serving European public administrations, businesses and ultimately the citizens. By doing so we are also contributing to European policies and supporting the Digital Decade.