Conference paper Open Access

Controlling Utterance Length in NMT-based Word Segmentation with Attention

Godard, Pierre; Besacier, Laurent; Yvon, François

DataCite XML Export

<?xml version='1.0' encoding='utf-8'?>
<resource xmlns:xsi="" xmlns="" xsi:schemaLocation="">
  <identifier identifierType="DOI">10.5281/zenodo.3524959</identifier>
      <creatorName>Godard, Pierre</creatorName>
      <affiliation>LIMSI, CNRS, Université Paris-Saclay, F 91 405 Orsay &amp; LIG, CNRS et Université Grenoble Alpes, F 38 000 Grenoble</affiliation>
      <creatorName>Besacier, Laurent</creatorName>
      <affiliation>LIG, CNRS et Université Grenoble Alpes, F 38 000 Grenoble</affiliation>
      <creatorName>Yvon, François</creatorName>
      <affiliation>LIMSI, CNRS, Université Paris-Saclay, F 91 405 Orsay</affiliation>
    <title>Controlling Utterance Length in NMT-based Word Segmentation with Attention</title>
    <date dateType="Issued">2019-11-02</date>
  <resourceType resourceTypeGeneral="Text">Conference paper</resourceType>
    <alternateIdentifier alternateIdentifierType="url"></alternateIdentifier>
    <relatedIdentifier relatedIdentifierType="DOI" relationType="IsVersionOf">10.5281/zenodo.3524958</relatedIdentifier>
    <relatedIdentifier relatedIdentifierType="URL" relationType="IsPartOf"></relatedIdentifier>
    <rights rightsURI="">Creative Commons Attribution 4.0 International</rights>
    <rights rightsURI="info:eu-repo/semantics/openAccess">Open Access</rights>
    <description descriptionType="Abstract">&lt;p&gt;One of the basic tasks of computational language documentation (CLD) is to identify word boundaries in an unsegmented phonemic stream. While several unsupervised monolingual word segmentation algorithms exist in the literature, they are challenged in real-world CLD settings by the small amount of available data. A possible remedy is to take advantage of glosses or translation in a foreign, well- resourced, language, which often exist for such data. In this paper, we explore and compare ways to exploit neural machine translation models to perform unsupervised boundary detection with bilingual information, notably introducing a new loss function for jointly learning alignment and segmentation. We experiment with an actual under-resourced language, Mboshi, and show that these techniques can effectively control the output segmentation length.&lt;/p&gt;</description>
All versions This version
Views 6863
Downloads 4040
Data volume 22.3 MB22.3 MB
Unique views 6459
Unique downloads 3838


Cite as