COMtext.SR: Serbian Language Resources for AI and NLP
Natural language processing is becoming an important foundation for digital services, automation and AI-based tools, but its development depends heavily on the availability of high-quality language resources. For Serbian, especially in both ekavica and ijekavica variants, such resources are still limited compared to widely used global languages. This creates a barrier for companies, public institutions and researchers who want to develop reliable Serbian-language applications, from document analysis and search tools to chatbots, decision-support systems and domain-specific AI solutions.
COMtext.SR addresses this gap by developing a basic set of publicly available resources and tools for the automatic processing of texts in the Serbian language. The project places special focus on professional domains that are highly relevant for digital transformation, but are still insufficiently covered by existing academic or commercial resources, such as legal, administrative, financial and medical texts. By supporting the processing of such content, the project contributes to the development of more advanced Serbian-language digital services and AI applications.
With that goal in mind, this project gathers and synchronizes the wider community (IT industry, academic community) that will contribute to the realization of this task by donating professional and material resources and intellectual property.
