SPEECH RECOGNITION: PROGRAMMING ASSIGNMENTS

This page contains pointers to the programming assignments used in this course. Submitting your solutions for grading involves the following:
I will grade your projects based on the clarity of your documentation.

Number Exercise(s) Description Due Date
01
IFC Programming
Learn to use the foundation classes to do basic signal processing. 01/17/00
02
Linear Prediction
Learn how to compute a parametric spectral model of a speech signal. 02/14/00
03
Principal Component Analysis
Classify data using a variety of statistical normalization techniques. 02/21/00
04
Mel Frequency-Spaced Cepstral Coefficients
Investigate the statistical properties of typical features used in speech recognition. 04/01/00
05
Digit Recognition
Investigate high performance speech recognition on a small vocabulary task. 04/08/00
06
Letter Recognition
Even small vocabulary tasks can prove to be quite challenging when words are acoustically similar. recognition. 04/15/00
04
Language Modeling
Use the SRI Language Modeling toolkit to train a trigram language model on the Switchboard transcriptions. 04/22/00
08
The Big Enchilada
Train and evaluate a context-dependent phone Switchboard speech recognition system. 04/29/00