@inproceedings{adelani-etal-2021-effect,
title = "The Effect of Domain and Diacritics in {Y}oruba{--}{E}nglish Neural Machine Translation",
author = "Adelani, David and
Ruiter, Dana and
Alabi, Jesujoba and
Adebonojo, Damilola and
Ayeni, Adesina and
Adeyemi, Mofe and
Awokoya, Ayodele Esther and
Espa{\~n}a-Bonet, Cristina",
editor = "Duh, Kevin and
Guzm{\'a}n, Francisco",
booktitle = "Proceedings of Machine Translation Summit XVIII: Research Track",
month = aug,
year = "2021",
address = "Virtual",
publisher = "Association for Machine Translation in the Americas",
url = "https://aclanthology.org/2021.mtsummit-research.6/",
pages = "61--75",
abstract = "Massively multilingual machine translation (MT) has shown impressive capabilities and including zero and few-shot translation between low-resource language pairs. However and these models are often evaluated on high-resource languages with the assumption that they generalize to low-resource ones. The difficulty of evaluating MT models on low-resource pairs is often due to lack of standardized evaluation datasets. In this paper and we present MENYO-20k and the first multi-domain parallel corpus with a especially curated orthography for Yoruba{--}English with standardized train-test splits for benchmarking. We provide several neural MT benchmarks and compare them to the performance of popular pre-trained (massively multilingual) MT models both for the heterogeneous test set and its subdomains. Since these pre-trained models use huge amounts of data with uncertain quality and we also analyze the effect of diacritics and a major characteristic of Yoruba and in the training data. We investigate how and when this training condition affects the final quality of a translation and its understandability.Our models outperform massively multilingual models such as Google ($+8.7$ BLEU) and Facebook M2M ($+9.1$) when translating to Yoruba and setting a high quality benchmark for future research."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="adelani-etal-2021-effect">
<titleInfo>
<title>The Effect of Domain and Diacritics in Yoruba–English Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Adelani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dana</namePart>
<namePart type="family">Ruiter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jesujoba</namePart>
<namePart type="family">Alabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Damilola</namePart>
<namePart type="family">Adebonojo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adesina</namePart>
<namePart type="family">Ayeni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mofe</namePart>
<namePart type="family">Adeyemi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayodele</namePart>
<namePart type="given">Esther</namePart>
<namePart type="family">Awokoya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cristina</namePart>
<namePart type="family">España-Bonet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of Machine Translation Summit XVIII: Research Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kevin</namePart>
<namePart type="family">Duh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francisco</namePart>
<namePart type="family">Guzmán</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Machine Translation in the Americas</publisher>
<place>
<placeTerm type="text">Virtual</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Massively multilingual machine translation (MT) has shown impressive capabilities and including zero and few-shot translation between low-resource language pairs. However and these models are often evaluated on high-resource languages with the assumption that they generalize to low-resource ones. The difficulty of evaluating MT models on low-resource pairs is often due to lack of standardized evaluation datasets. In this paper and we present MENYO-20k and the first multi-domain parallel corpus with a especially curated orthography for Yoruba–English with standardized train-test splits for benchmarking. We provide several neural MT benchmarks and compare them to the performance of popular pre-trained (massively multilingual) MT models both for the heterogeneous test set and its subdomains. Since these pre-trained models use huge amounts of data with uncertain quality and we also analyze the effect of diacritics and a major characteristic of Yoruba and in the training data. We investigate how and when this training condition affects the final quality of a translation and its understandability.Our models outperform massively multilingual models such as Google (+8.7 BLEU) and Facebook M2M (+9.1) when translating to Yoruba and setting a high quality benchmark for future research.</abstract>
<identifier type="citekey">adelani-etal-2021-effect</identifier>
<location>
<url>https://aclanthology.org/2021.mtsummit-research.6/</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>61</start>
<end>75</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Effect of Domain and Diacritics in Yoruba–English Neural Machine Translation
%A Adelani, David
%A Ruiter, Dana
%A Alabi, Jesujoba
%A Adebonojo, Damilola
%A Ayeni, Adesina
%A Adeyemi, Mofe
%A Awokoya, Ayodele Esther
%A España-Bonet, Cristina
%Y Duh, Kevin
%Y Guzmán, Francisco
%S Proceedings of Machine Translation Summit XVIII: Research Track
%D 2021
%8 August
%I Association for Machine Translation in the Americas
%C Virtual
%F adelani-etal-2021-effect
%X Massively multilingual machine translation (MT) has shown impressive capabilities and including zero and few-shot translation between low-resource language pairs. However and these models are often evaluated on high-resource languages with the assumption that they generalize to low-resource ones. The difficulty of evaluating MT models on low-resource pairs is often due to lack of standardized evaluation datasets. In this paper and we present MENYO-20k and the first multi-domain parallel corpus with a especially curated orthography for Yoruba–English with standardized train-test splits for benchmarking. We provide several neural MT benchmarks and compare them to the performance of popular pre-trained (massively multilingual) MT models both for the heterogeneous test set and its subdomains. Since these pre-trained models use huge amounts of data with uncertain quality and we also analyze the effect of diacritics and a major characteristic of Yoruba and in the training data. We investigate how and when this training condition affects the final quality of a translation and its understandability.Our models outperform massively multilingual models such as Google (+8.7 BLEU) and Facebook M2M (+9.1) when translating to Yoruba and setting a high quality benchmark for future research.
%U https://aclanthology.org/2021.mtsummit-research.6/
%P 61-75
Markdown (Informal)
[The Effect of Domain and Diacritics in Yoruba–English Neural Machine Translation](https://aclanthology.org/2021.mtsummit-research.6/) (Adelani et al., MTSummit 2021)
ACL