Domain-specialized models maximize access to unbiased, factual information concerning COVID-19 in different languages
SYSTRAN, the leader in AI-based translation technology, announces its partnership with TAUS, a pioneer in the language data space, to provide global access to secure and accurate translations specific to the coronavirus pandemic through the new Corona Crisis Translation Models.
“There are substantial volumes of information being produced and circulated about the virus, symptoms, new treatments, vaccines and data from all parts of the globe,” said CIO of SYSTRAN, John Paul Barraza. “Now more than ever, it is vital that information such as medical data, expert findings and guidelines are both readily available and accurate. However, translating this content accurately requires specific knowledge of medical and scientific terminology. By partnering with TAUS and building state-of-the-art models with SYSTRAN’s NMT technology we make translating this critical information as easy and accurate as possible.”
SYSTRAN created these models in 12 language combinations, across six languages, based on quality parallel data provided by TAUS. Together, SYSTRAN and TAUS are working to ensure that people and communities in need have access to accurate coronavirus-related information in their local language.
“Fighting the corona crisis is not the task of governments and public institutions alone, it is a battle that involves every citizen, every community and every industry globally,” said Director of TAUS, Jaap van der Meer. “We invited industry stakeholders to take part in this charity initiative that aims at making vital information accessible to people in their own languages. SYSTRAN’s contribution in providing production-ready Corona Crisis Translation Models has been uniquely significant as it helped put the collected data in use.”
The models are publicly available at no cost for text box translation on the SYSTRAN Translate website. The TAUS-owned models are available for commercial use via the SYSTRAN Marketplace, a catalog of specialized models for specific domains such as medical, healthcare, legal, finance, education, science, technology and more.