FI:PV158 Speech signal processing - Course Information
PV158 Speech signal processingFaculty of Informatics
- Extent and Intensity
- 2/1. 2 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: k (colloquium), z (credit).
- doc. Dr. Ing. Jan Černocký (lecturer), doc. RNDr. Ivan Kopeček, CSc. (deputy)
- Guaranteed by
- prof. PhDr. Karel Pala, CSc.
Department of Machine Learning and Data Processing - Faculty of Informatics
Contact Person: doc. RNDr. Ivan Kopeček, CSc.
- Thu 10:00–11:50 B007 and each odd Thursday 12:00–13:50 B117
- Course Enrolment Limitations
- The course is also offered to the students of the fields other than those the course is directly associated with.
- fields of study / plans the course is directly associated with
- there are 6 fields of study the course is directly associated with, display
- Course objectives
- Applications of speech processing, digital processing of speech signals, production and perception of speech, introduction to phonetics, pre-processing and basic parameters of speech, linear-predictive model, cepstrum, fundamental frequency estimation, coding - time domain and vocoders, recognition - DTW and HMM
- Informational contents of written and spoken form of speech.
- Techniques of signal processing applied to speech: Fourier transform, z-transform, linear filtering.
- Time domain and frequency domain behavoir of linear systems.
- Signal processing model of speech production.
- Excitation and filter.
- Determination of parameters using linear prediction.
- LPC coefficients and derived parameters (PARCOR, LAR,...).
- Speech analysis using short-time Fourier transform (STFT): filter-bank interpretation, computation using fast Fourier transform (FFT).
- Cepstral analysis.
- Parameterization with perceptually warped frequency axis.
- Fundamental frequency determination.
- Features for speech processing, criteria of choice.
- Measures of similarity between speech segments.
- Speech coding: waveform and parametric vocoders.
- Excitation modeling (CELP).
- Phonetic vocoders.
- Speech recognition: Hidden Markov Models (HMM).
- HMM training and HMM decoding.
- Extension of HMMs to continuous speech recognition.
- Statistical language models.
- The studied methods are experimentally exercised in computer laboratories (Matlab).
- Assessment methods (in Czech)
- tydne 2h prednaska. 2h pocitacovych cviceni 1x za 14 dni. Maly domaci projekt, presentace na posledni prednasce. Test v poc. laboratorich, pisemna zkouska.
- Language of instruction
- Further Comments
- The course is taught annually.
- Teacher's information