Conference paper Open Access
Philipp Wiesenbach; Stefan Riezler
<?xml version='1.0' encoding='utf-8'?> <resource xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://datacite.org/schema/kernel-4" xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4.1/metadata.xsd"> <identifier identifierType="DOI">10.5281/zenodo.3524924</identifier> <creators> <creator> <creatorName>Philipp Wiesenbach</creatorName> <affiliation>Computational Linguistics, Heidelberg University, Germany</affiliation> </creator> <creator> <creatorName>Stefan Riezler</creatorName> <affiliation>Computational Linguistics & IWR, Heidelberg University, Germany</affiliation> </creator> </creators> <titles> <title>Multi-Task Modeling of Phonographic Languages: Translating Middle Egyptian Hieroglyphs</title> </titles> <publisher>Zenodo</publisher> <publicationYear>2019</publicationYear> <dates> <date dateType="Issued">2019-11-02</date> </dates> <language>en</language> <resourceType resourceTypeGeneral="Text">Conference paper</resourceType> <alternateIdentifiers> <alternateIdentifier alternateIdentifierType="url">https://zenodo.org/record/3524924</alternateIdentifier> </alternateIdentifiers> <relatedIdentifiers> <relatedIdentifier relatedIdentifierType="DOI" relationType="IsVersionOf">10.5281/zenodo.3524923</relatedIdentifier> <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf">https://zenodo.org/communities/iwslt2019</relatedIdentifier> </relatedIdentifiers> <rightsList> <rights rightsURI="https://creativecommons.org/licenses/by/4.0/legalcode">Creative Commons Attribution 4.0 International</rights> <rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights> </rightsList> <descriptions> <description descriptionType="Abstract"><p>Machine translation of ancient languages faces a low-resource problem, caused by the limited amount of available textual source data&nbsp;and their translations. We present a multi-task modeling approach to translating Middle Egyptian that is inspired by recent successful&nbsp;approaches to multi-task learning in end-to-end speech translation. We leverage the phonographic aspect of the hieroglyphic writing&nbsp;system, and show that similar to multi-task learning of speech recognition and translation, joint learning and sharing of structural&nbsp;information between hieroglyph transcriptions, translations, and POS tagging can improve direct translation of hieroglyphs by several&nbsp;BLEU points, using a minimal amount of manual transcriptions.</p></description> </descriptions> </resource>
All versions | This version | |
---|---|---|
Views | 80 | 80 |
Downloads | 48 | 48 |
Data volume | 6.8 MB | 6.8 MB |
Unique views | 71 | 71 |
Unique downloads | 47 | 47 |