Title:

Speech Processing Systems

Code:SRE
Ac.Year:2009/2010
Term:Winter
Study plans:
ProgramBranchYearDuty
IT-MSC-2MBI-Elective
IT-MSC-2MBS-Elective
IT-MSC-2MGM2ndElective
IT-MSC-2MGM.2ndElective
IT-MSC-2MIN-Compulsory-Elective - group I
IT-MSC-2MIN.-Elective
IT-MSC-2MIS-Elective
IT-MSC-2MIS.-Elective
IT-MSC-2MMI-Elective
IT-MSC-2MMM-Elective
IT-MSC-2MPS-Elective
IT-MSC-2MPV-Elective
IT-MSC-2MSK-Elective
IT-MSC-2EITE2ndElective
Language:Czech, English
Public info:http://www.fit.vutbr.cz/study/courses/SRE/public/
Credits:5
Completion:examination (written)
Type of
instruction:
Hour/semLecturesSem. ExercisesLab. exercisesComp. exercisesOther
Hours:3900013
 ExaminationTestsExercisesLaboratoriesOther
Points:50150035
Guarantee:Černocký Jan, doc. Dr. Ing., DCGM
Lecturer:Burget Lukáš, Ing., Ph.D., DCGM
Černocký Jan, doc. Dr. Ing., DCGM
Fapšo Michal, Ing., DCGM
Glembek Ondřej, Ing., DCGM
Matějka Pavel, Ing., Ph.D., DCGM
Schwarz Petr, Ing., Ph.D., DCGM
Smrž Pavel, doc. RNDr., Ph.D., DCGM
Faculty:Faculty of Information Technology BUT
Department:Department of Computer Graphics and Multimedia FIT BUT
Prerequisites: 
Speech Signal Processing (ZRE), DCGM
 
Learning objectives:
To extend the on the structure of language (phonetics, phonology) and acquire bases of statistical classifiers. To get acquainted with advanced methods of speech recognition and coding. To get acquainted with advanced methods of language modeling and syntactic analysis.
Description:
Phonetics and phonology. Statistical pattern recognition. HMM training and adaptation. HMM recognition. Phoneme recognition. Keyword spotting and search. Speaker identification and verification. Language identification. CELP speech coding. Language modeling. Psycholinguistics. Probabilistic parsing.
Learning outcomes and competences:
Students will extend the knowledge acquired in the basic speech signal processing and natural language processing courses toward modern methods. They will get acquainted with methods currently deployed in industrial applications (GSM telephones or commercially available recognizers). They will get acquainted with promising methods existing in research environment.  They will deepen their knowledge of natural langugage processing and language modelilng. This course allows students to implement simple speech processing applications, as for example voice command of a process. However, first of all it enables them to join the development of complex systems for speech recognition and coding systems in both academic and industrial environments.
Syllabus of lectures:
  1. Phonetics and phonology - syllable structure, phonological processes and distinctive features.
  2. Statistical pattern classification I. - Bayesian framework, Maximum likelihood learning, Gaussian mixture models. Features for GMM modeling.
  3. Statistical pattern classification II. - Artificial Neural Networks, Support vector machines. Sequence modeling - Hidden Markov models. 
  4. HMM training and adaptation - MLLR, MAP, discriminative training.
  5. HMM recognition - pronunciation dictionaries and networks, language modeling, decoding, lattices.
  6. Phoneme recognition. Keyword spotting and search - LVCSR, acoustic and phonetic lattices. Figure of Merit.
  7. Speaker identification and verification - GMM, SVM. Channel normalization and compensation - feature mapping, eigen-voices and nuissance attributes projection (NAP). Evaluation of speaker verification: DET curves, EER, cost function.
  8. Language identification - acoustic vs. phonotactic, evaluation.
  9. Speech coding - CELP framework - adaptive and stochastic codebooks, GSM standards.
  10. Language modeling 1 - n-gram models, class-based models
  11. Language modeling 2 - language-specific features, factored-language models
  12. Psycholinguistics - word recognition models, word associations
  13. Probabilistic parsing - inside-outside algorithm, dependency parsing
Fundamental literature:
  1. Gussenhoven, J. and Jacobs, H.: Understanding Phonology, Oxford University Press, 1998, ISBN: 0-340-69218-9
  2. Psutka, J.: Komunikace s počítačem mluvenou řečí. Academia, Praha, 1995, ISBN 80-200-0203-0.
  3. Gold, B., Morgan, N.: Speech and audio signal processing, John Wiley & Sons, 2000, ISBN 0-471-35154-7.
  4. Moore, B.C.J.: An introduction to the psychology of hearing, Academic Press, 1989, ISBN 0-12-505627-3.
  5. Jelinek, F.: Statistical Methods for Speech Recognition, MIT Press, 1998, ISBN 0-262-10066-5.
  6. Manning, C. and Schütze, H.: Foundations of Statistical Natural Language Processing, MIT Press. Cambridge, MA: May 1999.
Study literature:
  1. Gold, B., Morgan, N.: Speech and audio signal processing, John Wiley & Sons, 2000, ISBN 0-471-35154-7.
Progress assessment:
  • mid-term test - 20pts
  • presentation of projects - 30pts
  • exam - 50pts