Computational Intelligence, SS08
2 VO 442.070 + 1 RU 708.070 last updated:
Course Notes (Skriptum)
Online Tutorials
Introduction to Matlab
Neural Network Toolbox
OCR with ANNs
Adaptive Filters
VC dimension
Gaussian Statistics
PCA, ICA, Blind Source Separation
Hidden Markov Models
Mixtures of Gaussians
Automatic Speech Recognition
Practical Course Slides
Animated Algorithms
Interactive Tests
Key Definitions
Literature and Links

Automatic Speech Recognition with HTK

A Tutorial for the Course Computational Intelligence


This tutorial introduces the main components for an automatic speech recognition system. The acoustic information is sampled as a signal suitable for processing by computers (a digital signal) and fed into a recognition process. The output of the system is a hypothesis for a transcription of the speech signal. The Hidden Markov Model Toolkit (HTK) [5] is used for building and manipulating hidden Markov models, being the core of most state-of-the-art speech recognition systems. HTK is used within this tutorial to build a simple speech recognizer.


To make full use of this tutorial you have to
  1. download the file which contains this tutorial (PDF and ps.gz) and an introduction of the HTK (PDF and ps.gz).
  2. The necessary HTK programs and data files are available from the homework assignment page.