@inproceedings{wu-etal-2019-transformer,
title = "Transformer-based Cascaded Multimodal Speech Translation",
author = "Wu, Zixiu and
Caglayan, Ozan and
Ive, Julia and
Wang, Josiah and
Specia, Lucia",
booktitle = "Proceedings of the 16th International Conference on Spoken Language Translation",
month = nov # " 2-3",
year = "2019",
address = "Hong Kong",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2019.iwslt-1.6",
abstract = "This paper describes the cascaded multimodal speech translation systems developed by Imperial College London for the IWSLT 2019 evaluation campaign. The architecture consists of an automatic speech recognition (ASR) system followed by a Transformer-based multimodal machine translation (MMT) system. While the ASR component is identical across the experiments, the MMT model varies in terms of the way of integrating the visual context (simple conditioning vs. attention), the type of visual features exploited (pooled, convolutional, action categories) and the underlying architecture. For the latter, we explore both the canonical transformer and its deliberation version with additive and cascade variants which differ in how they integrate the textual attention. Upon conducting extensive experiments, we found that (i) the explored visual integration schemes often harm the translation performance for the transformer and additive deliberation, but considerably improve the cascade deliberation; (ii) the transformer and cascade deliberation integrate the visual modality better than the additive deliberation, as shown by the incongruence analysis.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-etal-2019-transformer">
<titleInfo>
<title>Transformer-based Cascaded Multimodal Speech Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zixiu</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ozan</namePart>
<namePart type="family">Caglayan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Ive</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Josiah</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-nov" 2-3"</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 16th International Conference on Spoken Language Translation</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hong Kong</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes the cascaded multimodal speech translation systems developed by Imperial College London for the IWSLT 2019 evaluation campaign. The architecture consists of an automatic speech recognition (ASR) system followed by a Transformer-based multimodal machine translation (MMT) system. While the ASR component is identical across the experiments, the MMT model varies in terms of the way of integrating the visual context (simple conditioning vs. attention), the type of visual features exploited (pooled, convolutional, action categories) and the underlying architecture. For the latter, we explore both the canonical transformer and its deliberation version with additive and cascade variants which differ in how they integrate the textual attention. Upon conducting extensive experiments, we found that (i) the explored visual integration schemes often harm the translation performance for the transformer and additive deliberation, but considerably improve the cascade deliberation; (ii) the transformer and cascade deliberation integrate the visual modality better than the additive deliberation, as shown by the incongruence analysis.</abstract>
<identifier type="citekey">wu-etal-2019-transformer</identifier>
<location>
<url>https://aclanthology.org/2019.iwslt-1.6</url>
</location>
<part>
<date>2019-nov" 2-3"</date>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Transformer-based Cascaded Multimodal Speech Translation
%A Wu, Zixiu
%A Caglayan, Ozan
%A Ive, Julia
%A Wang, Josiah
%A Specia, Lucia
%S Proceedings of the 16th International Conference on Spoken Language Translation
%D 2019
%8 nov" 2 3"
%I Association for Computational Linguistics
%C Hong Kong
%F wu-etal-2019-transformer
%X This paper describes the cascaded multimodal speech translation systems developed by Imperial College London for the IWSLT 2019 evaluation campaign. The architecture consists of an automatic speech recognition (ASR) system followed by a Transformer-based multimodal machine translation (MMT) system. While the ASR component is identical across the experiments, the MMT model varies in terms of the way of integrating the visual context (simple conditioning vs. attention), the type of visual features exploited (pooled, convolutional, action categories) and the underlying architecture. For the latter, we explore both the canonical transformer and its deliberation version with additive and cascade variants which differ in how they integrate the textual attention. Upon conducting extensive experiments, we found that (i) the explored visual integration schemes often harm the translation performance for the transformer and additive deliberation, but considerably improve the cascade deliberation; (ii) the transformer and cascade deliberation integrate the visual modality better than the additive deliberation, as shown by the incongruence analysis.
%U https://aclanthology.org/2019.iwslt-1.6
Markdown (Informal)
[Transformer-based Cascaded Multimodal Speech Translation](https://aclanthology.org/2019.iwslt-1.6) (Wu et al., IWSLT 2019)
ACL