Introduction

Abstract

Automatic speech recognition has attracted various research activities since the 1950s. To achieve a high degree of user-friendliness, a natural and easy to use human-computer interface is targeted for technical applications. Since speech is the most important means of interpersonal communication, machines and computers can be operated more conveniently with the help of automated speech recognition and understanding [Furui, 2009].

The history of speech recognition and speaker identification is characterized by steady progress. Whereas in the 1950s the first realizations were mainly built on heuristic approaches, more sophisticated statistical techniques have been established and continuously developed since the 1980s [Furui, 2009]. Today a high level of recognition accuracy has been achieved which allows applications of increasing complexity to be controlled by speech.