Infopaket

Natural human-computer interfaces

ДВС.3.10

Вибіркова дисципліна для ОП

First

2022/2023

8 Semester

3

PLO 2. To use the modern mathematical apparatus of continuous and discrete analysis, linear algebra, analytical geometry, in professional activities to solve problems of a theoretical and applied nature in the process of designing and implementing informatization objects by industry. PLO 7. Be able to apply the methodology of simulation modeling of objects, processes and systems, plan and conduct experiments with models, make decisions about achieving the goal based on the results of simulation. PLO 21.3. Know artificial intelligence technologies and be able to apply them in solving practical problems. PLO 22.3. Know the technologies of implementation of "human-computer" interaction and be able to apply them.

1. Know the basic methods of processing, analysis and synthesis of voice, text, symbolic forms of information, methods of modeling complex systems, methods of data visualization, etc. 2. To be able to apply basic methods of modeling and recognition of communication information, creation of software for the study of voice information, and to master practical skills for solving this class of problems. 3. Have basic skills in information processing, programming, databases, algorithm development, pattern recognition.

The educational discipline includes the following sections. Methods of obtaining and processing voice information. Identification of the main characteristics in voice speech signals. Problems of educational samples. Phonetic characteristics of speech signals. Basic algorithms of concatenative and formant synthesis. Study of the human vocal tract, models of sound construction and propagation. Webster's equation for modeling the vocal tract. Basic prosodic characteristics of the human voice. Frequency filtering of signals. Methods based on energy characteristics of signals. Methods of selection of individual words. Methods of speech signal analysis. Linear prediction for speech synthesis and recognition. The main task is to provide knowledge on processing, modeling and recognition of voice and speech information, development of practical skills for solving educational and practical problems. It is taught in the 8th semester in the amount of 90 hours. (3 ECTS credits) in particular: lectures – 28 hours, consultations - 2 hours, independent work - 60 hours. 2 control papers and a credit are provided.

Basic: 1. Dzh.L.Flanagan. Analiz, sintez i vospriiatie rechi. Per. s angl. – M.: Sviaz-, 1968. 2. Becchetti C, Ricotti L.P. Speech recognition. Theory and C++ Implementation. John Wiley& Sons Ltd, 1999. 3. Jurafsky D., Martin J.H. Speech and Language Processing. An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition. Second Edition. – Pearson Prentice Hall, 2009. 4. Vintsiuk T. K. Analiz, raspoznavanie i interpretatsiia rechevykh signalov. – Kiev: Nauk. dumka, 1987. 5. Piegl L., Tiller W. The NURBS Book, – Berlin, Germany: Springer-Verlag, 1996. 6. Rabiner L., Juang B.-H. Fundamentals of Speech Recognition. – PTR Prentice Hall, 1993. ..

Lectures, consultations, independent work

Student evaluation forms: 1. Control work 1: РН1.1 – 30 p./18 p. 2. Control work 2: РН1.1, РН2.1 – 30 p./18 p. 3. Current assessment: RN 2.1, RN 4.1, RN 4.2 – 40 points. / 24 b. 7.2 Evaluation organization: On the basis of paragraph 4.1, 4.2 Regulations on the procedure for assessing students' knowledge and para. 4.6.1, 7.1.5, 7.1.11, 7.1.12 Regulations on the organization of the educational process: – the maximum number of points that can be obtained by a student during the semester – 100; - the credit is given based on the results of work in the semester as the sum of the points received for the control papers and the current assessment

Ukrainian