Introduction to Speech Recognition
Course schedule: Tuesdays 2-4pm in Ziskind Rm. 1, Thursdays 2-4pm in Ziskind Rm. 261
Course textbook: None (selected readings will be provided)
Instructor: Karen Livescu, firstname.lastname@example.org
Prerequisites: A good understanding of probability
This short course will introduce techniques used in speech technologies, mainly focusing on speech recognition. Speech recognition is one of the oldest and most complex structured sequence prediction tasks receiving significant research and commercial attention.
It is both a success story -- providing practical and commercially viable applications already -- and at the same time far from being solved. A number of ideas in statistical modeling and learning for sequences originated in speech recognition research.
It therefore provides a good case study for many of the techniques that are used in other areas of artificial intelligence. Sample topics to be covered: Acoustic signal processing, acoustic modeling with hidden
Markov models, pronunciation modeling, language modeling, and topics of current research interest (graphical models, adaptation to speaker/condition, discriminative approaches).