Wednesday, Oct. 22
09:15 - 09:30 Welcome, practical information
09:30 - 11:45 Phonetics (4h - Jacques/Wim) - sound examples(zip file), video
Basic phonetics (introduction, sound classes etc)
Spectrogram reading basics
Articulatory phonetics, sound production
Distinctive phonetic features, SPE, government phonology
Practical excercises
11:45 - 13:00 Lunch
13:00 - 14:00 Phonetics (cont)
14:15 - 16:00 Signal processing basics (2h, MHJ) additional slides
Discrete time representation, z-transform
Fourier representations, DFT
Digital filters
Filterbanks
Thursday, Oct 23
09:15 - 10:00 Estimation theory (basics) (1h, Marco)
10:15 - 11:45 Short time spectral estimation methods (2h, TSv)
Stochastic processes, power spectral densities
Non-parametric spectrum estimation
Parametric spectrum estimation
13:00 - 16:00 Models for speech analysis (3 1/2 h)
13:00 - 13:45 Speech models (1h, TSv)
LPC
PLP
MFCC
14:00 - 14:45 Noise compensation (1h, MHJ)
Spectral subtraction
Mean and variance normalization
15:00 - 15:30 Temporal represenations of speech (1/2 h, TSv)
ZCR
Short-time energy
15:45 - 16:30 Pitch and voicing estimation (1h, TSv)
Friday, Oct 24
09:15 - 10:00 Formant estimation (1h, TSv)
10:15 - 11:00 Auditory based methods for robust speech feature extraction (Bojana)
11:15 - 11:45 Statistical speech recognition
11:15 - 11:45 Theory basis (1h, MHJ)
Pattern recognition/classification basics
11:45 - 13:00 Lunch
13:15 - 15:00 Hidden Markov models (2h, MHJ)
Acoustic modeling
ML training
Discriminative training principles
15:15 - 15:45 Auditory models (TSv)
15:45 - 16:00 Summary and closing
Monday, Nov 24
09:15 - 09:30 Welcome, practical information
09:30 - 10:30 Linking human and automatic speech recognition research (Odette)
10:45 - 11:45 Units and lexica for automatic speech recognition (Ingunn)
11:45 - 13:00 Lunch
13:00 - 15:00 Artificial neural networks (2h, MHJ) additional figures
The perceptron
Feed-forward networks
Back-propagation training
ANN as classifier
ANN for posterior estimation
15:15 - 16:00 Graphical models - introduction (Marco)
Tuesday, Nov 25
09:15 - 11:00 Acoustic and lexical adaptation (Ingunn)
11:15 - 11:45 Language modeling (TSv)
11:45 - 13:00 Lunch
13:00 - 14:00 Language modeling (TSv, Line)
14:15 - 16:00 Graphical models (2h, Marco)
Bayesian networks and dynamical BN
Conditional random fields
Wednesday, Nov 26
09:15 - 11:45 Decoding (3h, TSv)
Finite state machines for automatic speech recognition (2h)
Theory basis
Finite state transducers
11:45 - 13:00 Lunch
13:00 - 13:30 Time and frequency domain techniques for phonetic feature detection
(Marco)
13:45 - 15:00 Performance evaluation (2h, TSv)
Evaluation principles
Significance evaluation
Designing tests
State-of-the-art in ASR
15:15 - 15:45 Summing up, practical information (TSv)