Research Problem:
This project is developing automatic speech analysis tools to
facilitate data mining of recordings of the oral arguments of the US
Supreme Court cases. The project will involve work with existing
speech tools and implementation of new methods of signal processing
for use in automatically identifying contentious discussions.
The student will work with Prof. Ostendorf and faculty from the
Linguistics department, together with a team of graduate and
undergraduate students. Challenges for an undergraduate researcher may
include: development of signal processing algorithms to characterize
speaking style and analysis of individual style variation, development
of classifiers for automatic detection of disfluent speech, and
implementation of tools to process large audio files.
Desired Skills/Background:
- Familiar with linux, Matlab and shell scripting
- Familiar with one or more of the following scripting languages:
perl, python, and ruby
- Good record in EE235 and EE341 preferred
For more information or to submit a resume, please contact Brian
Hutchinson (brianhutchinson@ee). Brian will meet with
all candidates, and he will select the most qualified to talk with
Prof. Ostendorf.
