Speech Reconstruction - Overview of State-of-the-art Systems

Abstract

This paper describes the speech reconstruction problem and lists the most frequent speaker errors. We show that deletion of these errors is not sufficient and more complex string operation should be done.

We overview state-of-the-art methods which try to solve this problem and present new ideas which we would like implement in the near future.

Keywords

• speech
• reconstruction
• overview
• state
• systems