SuperLectures.com

STRUCTURED OUTPUT LAYER NEURAL NETWORK LANGUAGE MODEL

Full Paper at IEEE Xplore

Language Modeling

Přednášející: Ilya Oparin, Autoři: Hai Son Le, LIMSI CNRS / Uni. Paris-Sud, France; Ilya Oparin, LIMSI CNRS, France; Alexandre Allauzen, LIMSI CNRS / Uni. Paris-Sud, France; Jean-Luc Gauvain, LIMSI CNRS, France; Francois Yvon, LIMSI CNRS / Uni. Paris-Sud, France

This paper introduces a new neural network language model (NNLM) based on word clustering to structure the output vocabulary: Structured Output Layer NNLM. This model is able to handle vocabularies of arbitrary size, hence dispensing with the design of short-lists that are commonly used in NNLMs. Several softmax layers replace the standard output layer in this model. The output structure depends on the word clustering which uses the continuous word representation induced by a NNLM. The GALE Mandarin data was used to carry out the speech-to-text experiments and evaluate the NNLMs. On this data the well tuned baseline system has a character error rate under 10%. Our model achieves consistent improvements over the combination of an n-gram model and classical short-list NNLMs both in terms of perplexity and recognition accuracy.


  Přepis řeči

|

  Slajdy

Zvětšit slajd | Zobrazit všechny slajdy

0:00:16

  1. slajd

0:00:27

  2. slajd

0:00:44

  3. slajd

0:01:21

  4. slajd

0:01:57

  5. slajd

0:02:16

  6. slajd

0:02:30

  7. slajd

0:02:42

  8. slajd

0:02:52

  9. slajd

0:03:01

 10. slajd

0:04:13

 11. slajd

0:04:25

 12. slajd

0:04:43

 13. slajd

0:05:52

 14. slajd

0:07:18

 15. slajd

0:08:20

 16. slajd

0:09:51

 17. slajd

0:10:41

 18. slajd

0:12:05

 19. slajd

0:12:16

 20. slajd

0:12:26

 21. slajd

0:13:09

 22. slajd

0:13:30

 23. slajd

0:14:49

 24. slajd

0:17:42

 25. slajd

0:19:12

     8. slajd

0:19:33

    12. slajd

  Komentáře

Please sign in to post your comment!

  Informace o přednášce

Nahráno: 2011-05-25 16:55 - 17:15, Club H
Přidáno: 9. 6. 2011 02:30
Počet zhlédnutí: 49
Rozlišení videa: 1024x576 px, 512x288 px
Délka videa: 0:22:23
Audio stopa: MP3 [7.65 MB], 0:22:23