Vedúci: Ing. Milan Rusko, PhD.
Research Focus
Theoretical Part
- Trainable (self-learning) automated speech processing systems
- Speech recognition in Slovak (automated system interface capable of understanding human speech)
- Speech synthesis in Slovak; legacy version
- Speaker recognition and verification (voice biometrics)
Application Part
- Development of speech recognizers for applications in telecommunications, robotics, medicine, banking, etc. The design of modern recognition systems requires the construction of speech training databases. Recognizers can be components of automated translation systems (machine translation) as well as input interfaces for voice-controlled remote telephone services.
- Development of speech synthesis systems with a higher degree of naturalness, achieving compatibility with widely used operating systems (Windows, Linux, Unix) and Java. Development of systems for the disabled and the visually impaired. Construction of specialized training databases for modern, high-level data-driven speech synthesis systems.
Additional Research Areas
- Phonetic research – SAMPA (Speech Assessment Methods Phonetic Alphabet) for the Slovak language
- Speech and musical acoustics
