SuperLectures.com

INFORMATIVE DIALECT RECOGNITION USING CONTEXT-DEPENDENT PRONUNCIATION MODELING

Language Identification

Full Paper at IEEE Xplore

Přednášející: Nancy Chen, Autoři: Nancy Chen, Massachusetts Institute of Technology, United States; Wade Shen, Joseph Campbell, Pedro Torres-Carrasquillo, MIT Lincoln Laboratory, United States

We propose an informative dialect recognition system that learns phonetic transformation rules, and uses them to identify dialects. A hidden Markov model is used to align reference phones with dialect-specific pronunciations to characterize when and how often substitutions, insertions, and deletions occur. Decision tree clustering is used to find context-dependent phonetic rules. We ran recognition tasks on 4 Arabic dialects. Not only do the proposed systems perform well on their own, but when fused with baselines they improve performance by 21-36% relative. In addition, our proposed decision-tree system beats the baseline monophone system in recovering phonetic rules by 21% relative. Pronunciation rules learned by our proposed system quantify the occurrence frequency of known rules, and suggest rule candidates for further linguistic studies.


  Přepis řeči

|

  Slajdy

Zvětšit slajd | Zobrazit všechny slajdy

0:00:16

  1. slajd

0:02:16

  2. slajd

0:04:24

  3. slajd

0:05:14

  4. slajd

0:06:07

  5. slajd

0:08:13

  6. slajd

0:08:43

  7. slajd

0:09:21

  8. slajd

0:09:51

  9. slajd

0:10:30

 10. slajd

0:10:57

 11. slajd

0:11:38

 12. slajd

0:12:05

 13. slajd

0:12:55

 14. slajd

0:13:23

 15. slajd

0:13:41

 16. slajd

0:14:14

 17. slajd

0:15:02

 18. slajd

  Komentáře

Please sign in to post your comment!

  Informace o přednášce

Nahráno: 2011-05-24 10:35 - 10:55, Panorama
Přidáno: 23. 6. 2011 17:40
Počet zhlédnutí: 56
Rozlišení videa: 1024x576 px, 512x288 px
Délka videa: 0:21:00
Audio stopa: MP3 [7.09 MB], 0:21:00