Language technology platform helps to enable multilingualism in the government's e-services and automate translation process for civil service.


Most Latvian public sector and e-service content was only provided in the official language of Latvia, which limited access to information for expats, foreign partners, etc. Additionally, since the information was only available in written form, it was not accessible to people with sight disabilities. Another pressing need for the government was to automate civil service translation workflow, as this requires a high volume of translation. Oftentimes, text translated by public administration is sensitive and confidential such as, for example, translations for procurement documentation, contracts, etc. Generic online translation service providers do not pay enough attention to data privacy and security. A translation platform owned by the state would manage the processed data, thus ensuring that sensitive data is not leaked. 


With all the aforementioned challenges faced by the Latvian public and civil service, the Latvian Government, represented by the Culture Information Systems Centre reached out to Tilde to develop a multi-language technology platform which had to resolve two challenges simultaneously – increase accessibility to public services for expats, foreign partners, and people with visual impairment, as well as help automate the translation process for the civil service.


Tilde developed – the Latvian state administration language technology platform that is freely accessible to every resident of Latvia. It offers three main functionalities: text translation, speech recognition (audio to text), and speech synthesis (text to audio). Via API, it is integrated into various state administration platforms and websites to facilitate multi-lingual access to e-administration data and e-services. In addition, it includes the management and digitalisation of language resources.

Since its launch, the platform has quickly become extremely popular with civil servants, ex-pats, foreign partners and all residents of Latvia. In the first three months, the platform was visited 267k times, which translates into approx. 3k visits per day. Overall, 352h or 14 days have been spent on pronunciation and 749h or 31 days on speech recognition. In turn, overall, 77 million words have been translated with the automated translation system, 2.8 million words have been spoken and 3.1 million words recognised during the three month is Latvia’s contribution to the European Union’s translation infrastructure, CEF.AT.

Benefits of are as follows:

  • Freely available to all citizens and residents
  • Customised for the Latvian language
  • Specially adapted to public sector terminology
  • Provides secure translations of data
  • API I for easy integration into government websites of public administration bodies
  • Continuously improved with user-provided edits and suggestions


Back to all case studies