nl-uiuc AT lists.cs.illinois.edu

Subject: Natural language research announcements

List archive

[nl-uiuc] Speech Recognition tools minicourse

From: "Mark Hasegawa-Johnson" <jhasegaw AT gmail.com>
To: nl-uiuc AT cs.uiuc.edu, nlp-lunch AT cs.uiuc.edu, "Mark H-J" <jhasegaw AT uiuc.edu>, sel AT ifp.iuc.edu, "Jont Allen" <jontalle AT illinois.edu>, jscole AT illinois.edu, "Chilin Shih" <cls AT uiuc.edu>, Sarah <sborys AT illinois.edu>, "Lae-Hoon Kim" <lkim9 AT uiuc.edu>, "Sharma, Harsh Vardhan" <hsharma AT uiuc.edu>, "Jui-Ting Huang" <jhuang29 AT illinois.edu>, "Bryce Lobdell" <lobdell AT illinois.edu>, "Yang Li" <yangli2 AT gmail.com>, "Yargop, Rahul Ulhas" <ryargop2 AT illinois.edu>, "harwath2 AT illinois.edu" <harwath2 AT illinois.edu>, "Dave Cohen" <americanski24 AT sbcglobal.net>, "Camille Goudeseune" <cog AT uiuc.edu>, "Dirk Bernhardt-Walther" <walther AT uiuc.edu>, "Thomas Huang" <huang AT ifp.uiuc.edu>, "Zhou, Xi" <xizhou2 AT uiuc.edu>, "Li, Zhen" <zhenli3 AT uiuc.edu>, "Jon Gunderson" <jongund AT illinois.edu>, "Marshall Scott Poole" <mspoole AT illinois.edu>, "Singer, Andrew Carl" <acsinger AT illinois.edu>, ymo AT uiuc.edu, "Chi Hu" <chihu1 AT illinois.edu>
Subject: [nl-uiuc] Speech Recognition tools minicourse
Date: Fri, 5 Dec 2008 09:04:20 -0600
List-archive: <http://lists.cs.uiuc.edu/pipermail/nl-uiuc>
List-id: Natural language research announcements <nl-uiuc.cs.uiuc.edu>

Hello everyone,

I will give a "Speech Recognition Tools" minicourse January 12-16, probably with a single lecture daily from 10:00-1:00. This course intends to teach the software tools used to train and test phone models, noise models, pronunciation models, and language models for automatic speech recognition. Emphasis will be on the practical use of publicly available software tools, including probably HTK, SRILM, GMTK, and OpenFST, and working with corpora that I have the right to redistribute widely, therefore probably AVICAR, ISLEX, UASPEECH, and wikipedia. The fundamental discrete-Markov statistical models (HMM, DBN, and FST) will be introduced as necessary to understand the tools. Content of the minicourse will be similar to a combination of the material in http://www.isle.uiuc.edu/courses/minicourse/index.html and http://www.isle.uiuc.edu/courses/htk/index.html. There will be no homework and no exam, but I will recommend tutorial assignments and provide sample code.

---
Mark Hasegawa-Johnson

[nl-uiuc] Speech Recognition tools minicourse, Mark Hasegawa-Johnson, 12/05/2008