公开(公告)号：CA2203132C

公开(公告)日：2004-11-16

申请号：CA2203132

申请日：1995-11-04

Applicant: IBM

Inventor： KUNZMANN SIEGFRIED ,  BANDARA UPALI ,  LEWIS BURN L ,  MOHR KARLHEINZ

IPC: G10L15/183 ,  G10L15/197 ,  G10L9/00 ,  G10L9/18

Abstract: Disclosed are a method and an apparatus for adapting, particularly reducing, the size of a language model, which comprises word n-grams, in a speech recognition system . The invention provides a mechanism to discard those n-grams for which the acoustic part of the system requires less support from the language model to recognize correctly. The proposed method is suitable for identifying those trigrams in a language model for the purpose of discarding during the built-time of the system. Provided is also another automatic classification scheme for words which allows the compression of a language model, but under retention of accuracy. Moreover it allows an efficient usage of sparsely available text corpora because even singleton trigrams are used when they are helpful. No additiona l software tools are needed to be developed because the main tool, the fast match scoring, is a module readily available in the known recognizers themselves. Further improvement of the method is accomplished by classification of words according to the common text in whic h they occur as far as they distinguish from each other acoustically. The invention opens the possibility to make speech recognition available in low-cost personal computers (PC's), even in portable computers like Laptops.

公开(公告)号：DE69517705T2

公开(公告)日：2000-11-23

申请号：DE69517705

申请日：1995-11-04

Applicant: IBM

Inventor： KUNZMANN SIEGFRIED ,  MOHR KARLHEINZ ,  BANDARA UPALI ,  LEWIS L

IPC: G10L15/183 ,  G10L15/197 ,  G10L15/00 ,  G10L15/22 ,  G10L15/28

Abstract: PCT No. PCT/EP95/04337 Sec. 371 Date Sep. 25, 1997 Sec. 102(e) Date Sep. 25, 1997 PCT Filed Nov. 4, 1995 PCT Pub. No. WO97/17694 PCT Pub. Date May 15, 1997In this speech recognition system, the size of the language model is reduced by discarding those n-grams that the acoustic part of the system can recognize most accurately without support from a language model. The n-grams can be discarded dynamically during the running of the system or during the build or setup-time of the system. Trigrams occurring infrequently in the text corpora are substituted for the discarded n-grams to increase the accuracy of the word recognitions.

公开(公告)号：CA2203132A1

公开(公告)日：1997-05-05

申请号：CA2203132

申请日：1995-11-04

Applicant: IBM

Inventor： BANDARA UPALI ,  KUNZMANN SIEGFRIED ,  LEWIS BURN L ,  MOHR KARLHEINZ

IPC: G10L15/183 ,  G10L15/197 ,  G10L9/00 ,  G10L9/18

Abstract: Disclosed are a method and an apparatus for adapting, particularly reducing, the size of a language model, which comprises word n-grams, in a speech recognition system. The invention provides a mechanism to discard those n-grams for which the acoustic part of the system requires less support from the language model to recognize correctly. The proposed method is suitable for identifying those trigrams in a language model for the purpose of discarding during the built-time of the system. Provided is also another automatic classification scheme for words which allows the compression of a language model, but under retention of accuracy. Moreover it allows an efficient usage of sparsely available text corpora because even singleton trigrams are used when they are helpful. No additional software tools are needed to be developed because the main tool, the fast match scoring, is a module readily available in the known recognizers themselves. Further improvement of the method is accomplished by classification of words according to the common text in which they occur as far as they distinguish from each other acoustically. The invention opens the possibility to make speech recognition available in low-cost personal computers (PC's), even in portable computers like Laptops.

公开(公告)号：DE4111781A1

公开(公告)日：1992-10-22

申请号：DE4111781

申请日：1991-04-11

Applicant: IBM

Inventor： BANDARA UPALI DIPL CHEM DR ,  MOHR KARLHEINZ ,  WALCH GEORG DIPL PHYS DR ,  KEPPEL ERIC DIPL PHYS DR ,  WOTHKE KLAUS DR ,  HITZENBERGER LUDWIG DR ,  SCHMIDT RUDOLF

IPC: G10L15/10 ,  G10L15/187

公开(公告)号：DE69517705D1

公开(公告)日：2000-08-03

申请号：DE69517705

申请日：1995-11-04

Applicant: IBM

Inventor： KUNZMANN SIEGFRIED ,  MOHR KARLHEINZ ,  BANDARA UPALI ,  LEWIS L

IPC: G10L15/183 ,  G10L15/197 ,  G10L15/00 ,  G10L15/22 ,  G10L15/28

Abstract: PCT No. PCT/EP95/04337 Sec. 371 Date Sep. 25, 1997 Sec. 102(e) Date Sep. 25, 1997 PCT Filed Nov. 4, 1995 PCT Pub. No. WO97/17694 PCT Pub. Date May 15, 1997In this speech recognition system, the size of the language model is reduced by discarding those n-grams that the acoustic part of the system can recognize most accurately without support from a language model. The n-grams can be discarded dynamically during the running of the system or during the build or setup-time of the system. Trigrams occurring infrequently in the text corpora are substituted for the discarded n-grams to increase the accuracy of the word recognitions.