The graduate of the course is familiar with deterministic and random signals, can perform signal filtering and understands voice technologies. The graduate is also able to solve practical tasks, i.e. choose and justify a suitable method and apply it.

Courses BPC-SI1, BPC-PP1, BPC-MA1, BPC-MA2 are required. Knowledge of the basics of systems and signal theory, mathematics at the bachelor's level and MATLAB.

Teaching methods include lectures and computer exercises. Student develops individual tasks during computer exercises. The course takes advantage of e-learning system (Moodle).

The conditions for successful completion of the course are specified in the annually updated announcement of the guarantor. The score is usually as follows:

- written test focused on counting examples max. 10 points,

- tasks in computer exercises max. 20 points,

- final exam max. 70 points.

1. Random signals, processes and their characteristics

2. Spectral parameters, windowing functions

3. Discrete linear systems

4. Linear signal filtering. FIR filters

5. IIR filters, designs of digital filters

6. Signal representation and classification of real phenomena

7. Transformation and optimization of signal features

8. Linear prediction of signals

9. Machinery speech recognition

10. Identification of persons by voice

11. Time transformations, time axis warping

12. Cepstral analysis of speech signals

13. Voice analysis for security purposes

All exercises are mandatory. Missed exercises must be made up by the end of the semester.

• Programme BKC-EKT Bachelor's 2 year of study, summer semester, compulsory