SDSU logo

Marie A. Roch
Associate Professor of Computer Science


CS 682 Speech Processing

Class meets: MW, 19:00-20:15, GMCS-254, Syllabus (PDF)

Note that in addition to normal office hours, you can see me on Wednesday mornings at 9:00 provided you send an e-mail the day before (make sure you do early enough that I have a chance to read it).

Text: Spoken Language Processing, Huang, Acero & Hon, Prentice Hall 2002.

Final Exam

The final exam for the course will be held in our normal classroom on Saturday, May 17th from 10:00 - 12:00. Bring a blue book (available at the SDSU book store).

Course materials

Tentative schedule for the semester.

Slides

  1. Motivation and architecture
  2. Sound, speech, and perception
  3. Digital systems, convolution demo
  4. Classifiers Part I, Part II
  5. Cepstral features
  6. Language modeling
  7. hidden Markov models (HMMs) Part I, Part II
  8. Decisions for acoustic modeling
  9. Search

Additional texts on speech recognition or audio

If you have difficulties with a presentation in Huang, Acero, and Hon, the simplest method is to simply ask me to explain it during office hours. If you so choose, you might want to consult another text. Here are a few relevant texts:

Frequently Asked Questions