The course outline will change regularly as the semester unfolds. Check regularly for updated readings and assignments.
| Dates |
Topic |
Reading |
Assignments |
| Sep 7 |
Overview of Speech and the Industry |
History of Automatic Speech Recognition,
Juang & Rabiner
State of the Art, Makhoul & Schwartz
|
Blog Assignment 1: Speech Application Review |
| Sep 12 |
Speech Recognizer Components |
Interview with Vlingo CTO Mike Phillips, Speech Tuning and Statistical Grammars |
|
| Sep 14 |
Building speech applications |
Speech Mashup Framework,
Speech Mashup Guide,
Kotelly, Writing Effective Prompts |
Speech application review due (postponed til links are up) |
| Sep 19 |
LVCSR Applications |
|
|
| Sep 21 |
Phonetics and dictionaries |
Jurafsky & Martin Ch. 7 |
|
| Sep 26 |
Front End Feature Extraction |
Jurafsky & Martin, SLP Ch. 9.1, 9.3
CMU Spectograms, Cepstrum, etc |
Blog Assignment 2: Speech Toolset Review
Quiz 1, Due Tuesday 9/26 5 pm |
| Sep 28 |
Viterbi Algorithm and HMM Review |
Jurafsky & Martin, Ch. 6, Ch.9.2 |
Dictionary confusability Due Oct 12 |
| Oct 3 |
Baum-Welch |
Eisner paper |
Eisner spreadsheet |
| Oct 5 |
Gaussians |
Jurafsky & Martin, Ch. 9.3-4 |
|
| Oct 10 |
Language Modeling |
Jurafsky & Martin, Ch. 4 (review)
Two Decades of LM, Rosenfeld |
Speech App Design description due |
| Oct 12 |
Ngrams & training |
Goodman, A bit of progress ... |
Phonetic confusability due |
| Oct 17 |
No class |
Modeling Conversational Speech,
Sentence Level Mixtures
Filtering Web Text
Linguistically Motivated Language Model
Integrating Symbolic and Statistical Approaches |
Quiz 2 questions |
| Oct 19 |
Decoding & Evaluation |
SLP Ch. 9.5-9 |
Quiz 2 due. Submit via Latte |
| Oct 24 |
Sphinx: hands on walk through |
SUN Technical Report & Sphinx Tutorial |
|
| Oct 26 |
Language Modeling Experiment |
|
|
| Oct 31 |
Overivew: HTK Speech Recognizer |
HTK Tutorial |
|
| Nov 2 |
DARPA Speech Evaluations |
|
Speech Analysis |
| Nov 7 |
LVCSR vs. Phonetics |
In class group discussions of speech papers |
Sphinx Decoding and Evaluation |
| Nov 9 |
Speech Recogition Analysis Presentations |
|
|
| Nov 14 |
Discussion of speech application prototypesf |
|
Working prototype speech application due |
| Nov 16 |
Speech Synthesis |
Jurafsky & Martin Ch. 8 |
|
| Nov 21 |
SPeech Synthesis |
|
|
| Nov 28 |
No class |
|
Sphinx LM Baslines due |
| Nov 30 |
Dialog |
Jurafsky & Martin Ch. 24 |
|
| Dec 5 |
Dialog |
PARADISE Dialog Evaluation |
Dialog Evaluation Blog Assignment |
| Dec 7 |
Voice User Interface Design, Guest speaker, Dr. Bernhard Suhm |
Human Factors in IVR - Chapter 1,
CHI2009 Suhm-Peterson |
Sphinx LM improvements (due 12/19) |
| Dec 12 |
Dialog Evaluation |
|
Dialog Evaluation Slide(s) due at 1 PM |
Dec 14
5 pm |
Final presentations and demonstrations |
|
Final appilcation presentations |
Dec 19 |
Sphinx LM improvements due |
|
|
