公开(公告)号：DE69315374T2

公开(公告)日：1998-05-28

申请号：DE69315374

申请日：1993-01-15

Applicant: IBM

Inventor： BROWN PETER FITZHUGH ,  DELLA PIETRA STEPHEN ANDREW ,  DELLA PIETRA VINCENT JOSEPH ,  MERCER ROBERT LEROY ,  JELINEK FREDERICK

IPC: G10L15/18 ,  G10L5/06 ,  G10L5/04 ,  G10L7/08 ,  G10L5/02

Abstract: A speech recognition system displays a source text of one or more words in a source language. The system has an acoustic processor for generating a sequence of coded representations of an utterance to be recognized. The utterance comprises a series of one or more words in a target language different from the source language. A set of one or more speech hypotheses, each comprising one or more words from the target language, are produced. Each speech hypothesis is modeled with an acoustic model. An acoustic match score for each speech hypothesis comprises an estimate of the closeness of a match between the acoustic model of the speech hypothesis and the sequence of coded representations of the utterance. A translation match score for each speech hypothesis comprises an estimate of the probability of occurrence of the speech hypothesis given the occurrence of the source text. A hypothesis score for each hypothesis comprises a combination of the acoustic match score and the translation match score. At least one word of one or more speech hypotheses having the best hypothesis scores is output as a recognition result.

公开(公告)号：ES2139690T3

公开(公告)日：2000-02-16

申请号：ES94111148

申请日：1994-07-18

Applicant: IBM

Inventor： BERGER ADAM L ,  BROWN PETER FITZHUGH ,  DELLA PIETRA STEPHEN ANDREW ,  KEHLER ANDREW SCOTT ,  MERCER ROBERT LEROY ,  DELLA PIETRA VINCENT JOSEPH

IPC: G06F17/28

公开(公告)号：DE69315374D1

公开(公告)日：1998-01-08

申请号：DE69315374

申请日：1993-01-15

Applicant: IBM

Inventor： BROWN PETER FITZHUGH ,  DELLA PIETRA STEPHEN ANDREW ,  DELLA PIETRA VINCENT JOSEPH ,  MERCER ROBERT LEROY ,  JELINEK FREDERICK

IPC: G10L15/18 ,  G10L5/06 ,  G10L5/04 ,  G10L7/08 ,  G10L5/02

Abstract: A speech recognition system displays a source text of one or more words in a source language. The system has an acoustic processor for generating a sequence of coded representations of an utterance to be recognized. The utterance comprises a series of one or more words in a target language different from the source language. A set of one or more speech hypotheses, each comprising one or more words from the target language, are produced. Each speech hypothesis is modeled with an acoustic model. An acoustic match score for each speech hypothesis comprises an estimate of the closeness of a match between the acoustic model of the speech hypothesis and the sequence of coded representations of the utterance. A translation match score for each speech hypothesis comprises an estimate of the probability of occurrence of the speech hypothesis given the occurrence of the source text. A hypothesis score for each hypothesis comprises a combination of the acoustic match score and the translation match score. At least one word of one or more speech hypotheses having the best hypothesis scores is output as a recognition result.

公开(公告)号：DE69230871D1

公开(公告)日：2000-05-11

申请号：DE69230871

申请日：1992-07-10

Applicant: IBM

Inventor： BROWN PETER FITZHUGH ,  COCKE JOHN ,  DELLA PIETRA STEPHEN ANDREW ,  DELLA PIETRA VINCENT JOSEPH ,  JELINEK FREDERICK ,  LAI JENNIFER CEIL ,  MERCER ROBERT LEROY

IPC: G10L15/14 ,  G06F17/27 ,  G06F17/28 ,  G10L15/00

公开(公告)号：AT187564T

公开(公告)日：1999-12-15

申请号：AT94111148

申请日：1994-07-18

Applicant: IBM

Inventor： BERGER ADAM L ,  BROWN PETER FITZHUGH ,  DELLA PIETRA STEPHEN ANDREW ,  KEHLER ANDREW SCOTT ,  MERCER ROBERT LEROY ,  DELLA PIETRA VINCENT JOSEPH

IPC: G06F17/28