Module status: ready
This module examines what should be contained in the pre-recorded natural speech database. The focus of the videos is on a database for unit selection synthesis. The class will discuss what changes as we move towards state-of-the-art methods.
You should already have a good understanding of:
- How the front end works (covered in Speech Processing) and what the linguistic specification contains
- Automatic Speech Recognition using Hidden Markov models, to the level covered in Speech Processing