Automatic Speech Recognition

"Automatic Speech Recognition" is an introduction to the field of automatic speech recognition. The course starts with the basics of human speech production and perception as well as theory of signal processing. The main part of the course covers how speech can be modeled for automatic speech recognition and the state-of-the-art in automatic speech recognition. The course concludes with approaches and methods for using automatic speech recognition in human-machine communication in various contexts.
Language

Lecture slides are in English. The lecture is in a mix of German and English. Tutorials and exercise sheets are English only.

Participants need to be able to work in German.
Aims

    Gain an overview over human speech production and perception
    Know the theoretical foundations of signal processing for automatic speech recognition
    Know the theoretical foundations of speech modeling with a focus on statistical modeling
    Know the components and building blocks of an automatic speech recognition system
    Be able to assess the potential and boundaries of modern speech recognition technology