SuperLectures.com

ACOUSTIC MODELS IN KALDI

Presented by: Arnab Ghoshal, Author(s): Arnab Ghoshal

This talk introduces the Kaldi speech recognition toolkit: a new speech recognition toolkit written in C++ that uses FSTs for training and testing. It uses the OpenFst library and links against BLAS and LAPACK for linear algebra support. The aim is to create a clean, flexible and well-structured toolkit for speech recognition researchers.

  Speech Transcript

  Slides

Enlarge the slide | Show all slides in a pop-up window

0:00:16

  1. slide

0:00:41

  2. slide

0:02:36

  3. slide

0:03:52

  4. slide

0:04:18

  5. slide

0:04:55

  6. slide

0:05:32

  7. slide

0:06:25

  8. slide

0:07:26

  9. slide

0:09:06

 10. slide

0:09:48

 11. slide

  Comments

Please sign in to post your comment!

  Lecture Information