Learning a translation model from word lattices

Oliver Adams, Graham Neubig, Trevor Cohn, Steven Bird

Research output: Chapter in Book/Report/Conference proceedingConference Paper published in Proceedingspeer-review

Abstract

Translation models have been used to improve automatic speech recognition when speech input is paired with a written translation, primarily for the task of computer-aided translation. Existing approaches require large amounts of parallel text for training the translation models, but for many language pairs this data is not available. We propose a model for learning lexical translation parameters directly from the word lattices for which a transcription is sought. The model is expressed through composition of each lattice with a weighted finite-state transducer representing the translation model, where inference is performed by sampling paths through the composed finite-state transducer. We show consistent word error rate reductions in two datasets, using between just 20 minutes and 4 hours of speech input, additionally outperforming a translation model trained on the 1-best path.

Original languageEnglish
Title of host publicationProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
PublisherInternational Speech and Communication Association
Pages2518-2522
Number of pages5
DOIs
Publication statusPublished - 1 Jan 2016
Externally publishedYes
Event17th Annual Conference of the International Speech Communication Association, INTERSPEECH 2016 - San Francisco, United States
Duration: 8 Sep 201616 Sep 2016

Publication series

NameProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
ISSN (Print)2308-457X

Conference

Conference17th Annual Conference of the International Speech Communication Association, INTERSPEECH 2016
Country/TerritoryUnited States
CitySan Francisco
Period8/09/1616/09/16

Fingerprint

Dive into the research topics of 'Learning a translation model from word lattices'. Together they form a unique fingerprint.

Cite this