Speech and Language Processing

-Introduction to Automatic Speech Recognition-

Lecture material for TAIST-Tokyo Tech Program

Lecture:

1. Speech recognition based on GMM, HMM, and N-gram pdf
   Lecture: 2021/10/4 13:00-16:00 ICT (15:00-18:00 JST)
2. Maximum likelihood estimation and EM algorithm pdf
   Lecture: 2021/10/4 13:00-16:00 ICT (15:00-18:00 JST)
3. Bayesian network and Bayesian inference pdf
   Lecture: 2021/10/5 13:00-16:00 ICT (15:00-18:00 JST)
4. Variational inference and sampling pdf
   Lecture: 2021/10/5 13:00-16:00 ICT (15:00-18:00 JST)
5. Neural network based acoustic and language models pdf
   Lecture: 2021/10/6 13:00-16:00 ICT (15:00-18:00 JST)
6. Weighted finite state transducer (WFST) and speech decoding pdf
   Lecture: 2021/10/6 13:00-16:00 ICT (15:00-18:00 JST)

*sound is not available in the PDF.
*Attendance/Every day exercise: Explained in the lectures.

Takahiro Shinozaki
Department of Information and Communication Engineering
School of Engineering
Tokyo Institute of Technology

---

[back to home]