Algorithms for statistical translation of spoken language

Herrmann Ney, Sonja Nießen, Franz Josef Och, Hassan Sawaf, Christoph Tillmann, Stephan Vogel

Research output: Contribution to journalArticle

49 Citations (Scopus)

Abstract

In this paper, we describe three approaches to statistical translation and present experimental results. The statistical translation approach uses two types of information: a translation model and a language model. The language model used is a bigram or general m-gram model. The translation model is decomposed into a lexical model and an alignment model. There are three approaches that are presented and tested in detail: the quasi-monotone alignment approach, the inverted alignment approach, and the alignment template approach. For each of these three approaches, a suitable search method is presented. The system has been tested on a limited-domain spoken-language task for which a bilingual corpus is available: the Verbmobil task (German-English, 7000-word vocabulary). We present experimental results for each of the three approaches. The experimental tests were performed on both the text transcription and the speech recognizer output.

Original languageEnglish
Pages (from-to)24-36
Number of pages13
JournalIEEE Transactions on Speech and Audio Processing
Volume8
Issue number1
DOIs
Publication statusPublished - 1 Jan 2000
Externally publishedYes

Fingerprint

alignment
Transcription
templates
output

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Algorithms for statistical translation of spoken language. / Ney, Herrmann; Nießen, Sonja; Och, Franz Josef; Sawaf, Hassan; Tillmann, Christoph; Vogel, Stephan.

In: IEEE Transactions on Speech and Audio Processing, Vol. 8, No. 1, 01.01.2000, p. 24-36.

Research output: Contribution to journalArticle

Ney, Herrmann ; Nießen, Sonja ; Och, Franz Josef ; Sawaf, Hassan ; Tillmann, Christoph ; Vogel, Stephan. / Algorithms for statistical translation of spoken language. In: IEEE Transactions on Speech and Audio Processing. 2000 ; Vol. 8, No. 1. pp. 24-36.
@article{e9fc2ccc58da47418837a1401426a95e,
title = "Algorithms for statistical translation of spoken language",
abstract = "In this paper, we describe three approaches to statistical translation and present experimental results. The statistical translation approach uses two types of information: a translation model and a language model. The language model used is a bigram or general m-gram model. The translation model is decomposed into a lexical model and an alignment model. There are three approaches that are presented and tested in detail: the quasi-monotone alignment approach, the inverted alignment approach, and the alignment template approach. For each of these three approaches, a suitable search method is presented. The system has been tested on a limited-domain spoken-language task for which a bilingual corpus is available: the Verbmobil task (German-English, 7000-word vocabulary). We present experimental results for each of the three approaches. The experimental tests were performed on both the text transcription and the speech recognizer output.",
author = "Herrmann Ney and Sonja Nie{\ss}en and Och, {Franz Josef} and Hassan Sawaf and Christoph Tillmann and Stephan Vogel",
year = "2000",
month = "1",
day = "1",
doi = "10.1109/89.817451",
language = "English",
volume = "8",
pages = "24--36",
journal = "IEEE Transactions on Audio, Speech and Language Processing",
issn = "1558-7916",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Algorithms for statistical translation of spoken language

AU - Ney, Herrmann

AU - Nießen, Sonja

AU - Och, Franz Josef

AU - Sawaf, Hassan

AU - Tillmann, Christoph

AU - Vogel, Stephan

PY - 2000/1/1

Y1 - 2000/1/1

N2 - In this paper, we describe three approaches to statistical translation and present experimental results. The statistical translation approach uses two types of information: a translation model and a language model. The language model used is a bigram or general m-gram model. The translation model is decomposed into a lexical model and an alignment model. There are three approaches that are presented and tested in detail: the quasi-monotone alignment approach, the inverted alignment approach, and the alignment template approach. For each of these three approaches, a suitable search method is presented. The system has been tested on a limited-domain spoken-language task for which a bilingual corpus is available: the Verbmobil task (German-English, 7000-word vocabulary). We present experimental results for each of the three approaches. The experimental tests were performed on both the text transcription and the speech recognizer output.

AB - In this paper, we describe three approaches to statistical translation and present experimental results. The statistical translation approach uses two types of information: a translation model and a language model. The language model used is a bigram or general m-gram model. The translation model is decomposed into a lexical model and an alignment model. There are three approaches that are presented and tested in detail: the quasi-monotone alignment approach, the inverted alignment approach, and the alignment template approach. For each of these three approaches, a suitable search method is presented. The system has been tested on a limited-domain spoken-language task for which a bilingual corpus is available: the Verbmobil task (German-English, 7000-word vocabulary). We present experimental results for each of the three approaches. The experimental tests were performed on both the text transcription and the speech recognizer output.

UR - http://www.scopus.com/inward/record.url?scp=0033874696&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033874696&partnerID=8YFLogxK

U2 - 10.1109/89.817451

DO - 10.1109/89.817451

M3 - Article

AN - SCOPUS:0033874696

VL - 8

SP - 24

EP - 36

JO - IEEE Transactions on Audio, Speech and Language Processing

JF - IEEE Transactions on Audio, Speech and Language Processing

SN - 1558-7916

IS - 1

ER -