Keeletehnoloogia

Keeletehnoloogia (language technology) is the interdisciplinary field that studies how to model, analyze, and generate human language with computers. It encompasses natural language processing, speech technology, and text-oriented applications, as well as the development of linguistic resources and tools. The goal is to enable computers to understand, interpret, and produce language in a way that is useful for humans.

Key areas include natural language processing (parsing, tagging, information extraction, machine translation), speech technology (automatic speech recognition, speech synthesis), and interaction technologies such as dialogue systems and chatbots. It also includes text analytics, sentiment analysis, question answering, and handwriting recognition. Methods rely on computational linguistics, statistics, and machine learning, increasingly with deep learning, and depend on language resources such as annotated corpora, lexicons, and grammars.

Processing Estonian and other languages requires language-specific approaches due to features like rich morphology, compounding, and syntax. Morphological analyzers, lemmatizers, POS taggers, and dependency parsers are commonly used, along with cross-linguistic tools and standards like Universal Dependencies.

Resources and collaboration are essential, with research institutions, universities, and industry contributing corpora, benchmarks, and open-source software. European and global collaborations help develop tools for less-resourced languages, including Estonian, enabling digital services, education, accessibility, and government administration.

Applications include voice assistants, translation services, search and information retrieval, accessibility tools, and multilingual digital services. Ongoing work focuses on improving accuracy, robustness, and fairness, and on expanding language coverage.