Language Technology for Multilingual Europe: An Analysis of a Large-Scale Survey regarding Challenges, Demands, Gaps and Needs

Language Technology for Multilingual Europe: An Analysis of a Large-Scale Survey regarding Challenges, Demands, Gaps and Needs Georg Rehm, Stefanie Hegele DFKI GmbH, Alt-Moabit 91c, 10559 Berlin, Germany Corresponding author: georg.rehm@dfki.de Abstract We present the analysis of a large-scale survey titled “Language Technology for Multilingual Europe”, conducted between May and June 2017. A total of 634 participants in 52 countries responded to the survey. Its main purpose was to collect input, feedback and ideas from the European Language Technology research and innovation community in order to assess the most prominent research areas, projects and applications, but, more importantly to identify the biggest challenges, obstacles and gaps Europe is currently facing with regard to its multilingual setup and technological solutions. Participants were encouraged to share concrete suggestions and recommendations on how present challenges can be turned into opportunities in the context of a potential long-term, large-scale, Europe-wide research, development and innovation funding programme, currently titled Human Language Project. Keywords: Multilinguality, LR National/International Projects, Infrastructural/Policy Issues, LR Infrastructures and Architectures 1. Introduction Europe is a multilingual society with 24 official Member State languages and many additional unofficial and regional languages as well as languages of minorities, immigrants and important trade partners. Nevertheless, day in and day out, language barriers keep severely hampering the free flow of information, thought, ideas, goods and products through the continent. Powerful multilingual as well as cross-lingual and monolingual language technologies, mak- ing use of the latest Artificial Intelligence algorithms in combination with ever-growing data sets, have the potential of helping to overcome language barriers. The recent study “Language Equality in the Digital Age – Towards a Human Language Project”, commissioned by the European Parliament’s Science and Technology Options Assessment Committee (STOA), recommends, to the Euro- pean Union, to initiate a new, large-scale European Lan- guage Technology research, development and innovation flagship programme, called, in the study, the Human Lan- guage Project (HLP) (STOA, 2017). It is foreseen to be a long-term European collaborative programme between re- search, innovation, industry, academia, administrations and citizens with the goal of achieving the next scientific break- throughs for the automatic processing and generation of written or spoken natural language. In addition to basic research, the Human Language Project1 is foreseen to in- clude applied research as well as innovation and commer- cialisation activities. Important research themes are, among others, (1) Crosslingual Big Data Language Analytics, (2) High-Quality Machine Translation, (3) Meaning, Semantics and Knowledge as well as (4) Conversational Technologies. 1While identical in name, the “Human Language Project” only bears a marginal relationship to previous initiatives bearing the same name. Among others, Abney and Bird (Abney and Bird, 2010) called their “universal corpus of the world’s languages” the “Human Language Project”. In 2012 TAUS launched their con- cept for an open language resources and tools platform, which TAUS called “Human Language Project”, (see:https://www.taus. net/knowledgebase/index.php/Human_Language_Project). The initiative specified in the STOA report has a much broader scope and set of objectives than the two initiatives mentioned above. A key goal of our …

Content truncated to 500 words