Show simple item record

dc.contributor.authorWang, Yuen
dc.contributor.authorWong, JHMen
dc.contributor.authorGales, Marken
dc.contributor.authorKnill, Katherineen
dc.contributor.authorRagni, Antonen
dc.date.accessioned2019-01-12T00:30:24Z
dc.date.available2019-01-12T00:30:24Z
dc.date.issued2019-02-11en
dc.identifier.isbn9781538643341en
dc.identifier.issn2639-5479
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/287878
dc.description.abstractA high performance automatic speech recognition (ASR) system is an important constituent component of an automatic language assessment system for free speaking language tests. The ASR system is required to be capable of recognising non-native spontaneous English speech and to be deployable under real-time conditions. The performance of ASR systems can often be significantly improved by leveraging upon multiple systems that are complementary, such as an ensemble. Ensemble methods, however, can be computationally expensive, often requiring multiple decoding runs, which makes them impractical for deployment. In this paper, a lattice-free implementation of sequence-level teacher-student training is used to reduce this computational cost, thereby allowing for real-time applications. This method allows a single student model to emulate the performance of an ensemble of teachers, but without the need for multiple decoding runs. Adaptations of the student model to speakers from different first languages (L1s) and grades are also explored.
dc.description.sponsorshipCambridge Assessment English
dc.titleSequence Teacher-Student Training of Acoustic Models for Automatic Free Speaking Language Assessmenten
dc.typeConference Object
prism.endingPage1000
prism.publicationDate2019en
prism.publicationName2018 IEEE Spoken Language Technology Workshop, SLT 2018 - Proceedingsen
prism.startingPage994
dc.identifier.doi10.17863/CAM.35192
dcterms.dateAccepted2018-09-03en
rioxxterms.versionofrecord10.1109/SLT.2018.8639557en
rioxxterms.versionAM
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserveden
rioxxterms.licenseref.startdate2019-02-11en
dc.contributor.orcidGales, Mark [0000-0002-5311-8219]
dc.contributor.orcidKnill, Katherine [0000-0003-1292-2769]
rioxxterms.typeConference Paper/Proceeding/Abstracten
pubs.funder-project-idCambridge Assessment (unknown)
cam.orpheus.successThu Nov 05 11:53:18 GMT 2020 - Embargo updated*
rioxxterms.freetoread.startdate2020-02-11


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record