Book Image

Rapid BeagleBoard Prototyping with MATLAB and Simulink

Book Image

Rapid BeagleBoard Prototyping with MATLAB and Simulink

Overview of this book

As an open source embedded single-board computer with many standard interfaces, Beagleboard is ideal for building embedded audio/video systems to realize your practical ideas. The challenge is how to design and implement a good digital processing algorithm on Beagleboard quickly and easily without intensive low-level coding. Rapid BeagleBoard Prototyping with MATLAB and Simulink is a practical, hands-on guide providing you with a number of clear, step-by-step exercises which will help you take advantage of the power of Beagleboard and give you a good grounding in rapid prototyping techniques for your audio/video applications. Rapid BeagleBoard Prototyping with MATLAB and Simulink looks at rapid prototyping and how to apply these techniques to your audio/video applications with Beagleboard quickly and painlessly without intensive manual low-level coding. It will take you through a number of clear, practical recipes that will help you to take advantage of both the Beagleboard hardware platform and Matlab/Simulink signal processing. We will also take a look at building S-function blocks that work as hardware drivers and interfaces for Matlab/Simulink. This gives you more freedom to explore the full range of advantages provided by Beagleboard. By the end of this book, you will have a clear idea about Beagleboard and Matlab/Simulink rapid prototyping as well as how to develop voice recognition systems, motion detection systems with I/O access, and serial communication for your own applications such as a smart home.
Table of Contents (15 chapters)
Rapid BeagleBoard Prototyping with MATLAB and Simulink
Credits
About the Authors
About the Reviewers
www.PacktPub.com
Preface
Index

Structure of a voice recognition system


Although voice recognition systems differ in various ways, many of them, if not all, require some form of training which acclimatizes the system. Voice recognition systems usually share some common and fundamental techniques and have a similar structure as shown in the following figure:

A voice recognition system usually consists of three function blocks:

  • Feature extraction

  • Pattern analysis (for training)

  • Pattern matching (for recognition)

The workflow of a voice recognition system can be divided into two sessions:

  • Training

  • Recognition

Feature extraction is a process of transforming the large number of audio samples into a relatively small set of discriminatory features. The features are carefully chosen so that the features set are expected to be a good representation of the voice, without using the whole and large amount of audio data samples. Pattern analysis is to find the distribution of these features and their relationship with the meanings of the...