Spoken Language Technology. Speaker Recognition
Coordinated by: Corneliu Burileanu
The course aims to provide students with basic knowledge about speech signal processing and speech and speaker recognition. Algorithms for speech signal analysis and synthesis are studied, including the parametric representation of the signal, the extraction of speech signal parameters in the time and frequency domains, and compression techniques. It also studies the characteristics of digital signal processors and their use in spoken language technology. An important part of the course deals with speech and speaker recognition; the most effective methods in the field are being discussed, emphasizing the particularities of speech processing in Romanian.