Lab 2: HMM's and You

ELEN E6884/COMS 86884: Speech Recognition

Due: October 17, 2005 at 12:01am

Table of Contents

1. Overview

2. Part 1: Implement the Viterbi algorithm and Gaussian likelihood evaluation

2.1. Background
2.2. The Decoder
2.3. The Big HMM Graph
2.4. The Dynamic Programming Chart
2.5. What You're Supposed to Do
2.6. Compiling and testing

3. Part 2: Implement most of the Forward-Backward algorithm

3.1. The Plan
3.2. What You're Supposed to Do

4. Part 3: Implement Gaussian training within the Forward-Backward algorithm

5. Part 4: Train a model from scratch, and evaluate it on various digit test sets

6. What is to be handed in

		Next
		Overview