Low Resource Multimodal Neural Machine Translation of English-Hindi in News Domain

Proceedings of the FirstWorkshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021) Pub Date : 1900-01-01 DOI:10.26615/978-954-452-073-1_004

Loitongbam Sanayai Meetei, Thoudam Doren Singh, Sivaji Bandyopadhyay

{"title":"Low Resource Multimodal Neural Machine Translation of English-Hindi in News Domain","authors":"Loitongbam Sanayai Meetei, Thoudam Doren Singh, Sivaji Bandyopadhyay","doi":"10.26615/978-954-452-073-1_004","DOIUrl":null,"url":null,"abstract":"Incorporating multiple input modalities in a machine translation (MT) system is gaining popularity among MT researchers. Unlike the publicly available dataset for Multimodal Machine Translation (MMT) tasks, where the captions are short image descriptions, the news captions provide a more detailed description of the contents of the images. As a result, numerous named entities relating to specific persons, locations, etc., are found. In this paper, we acquire two monolingual news datasets reported in English and Hindi paired with the images to generate a synthetic English-Hindi parallel corpus. The parallel corpus is used to train the English-Hindi Neural Machine Translation (NMT) and an English-Hindi MMT system by incorporating the image feature paired with the corresponding parallel corpus. We also conduct a systematic analysis to evaluate the English-Hindi MT systems with 1) more synthetic data and 2) by adding back-translated data. Our finding shows improvement in terms of BLEU scores for both the NMT (+8.05) and MMT (+11.03) systems.","PeriodicalId":114625,"journal":{"name":"Proceedings of the FirstWorkshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the FirstWorkshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.26615/978-954-452-073-1_004","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 4

Abstract

Incorporating multiple input modalities in a machine translation (MT) system is gaining popularity among MT researchers. Unlike the publicly available dataset for Multimodal Machine Translation (MMT) tasks, where the captions are short image descriptions, the news captions provide a more detailed description of the contents of the images. As a result, numerous named entities relating to specific persons, locations, etc., are found. In this paper, we acquire two monolingual news datasets reported in English and Hindi paired with the images to generate a synthetic English-Hindi parallel corpus. The parallel corpus is used to train the English-Hindi Neural Machine Translation (NMT) and an English-Hindi MMT system by incorporating the image feature paired with the corresponding parallel corpus. We also conduct a systematic analysis to evaluate the English-Hindi MT systems with 1) more synthetic data and 2) by adding back-translated data. Our finding shows improvement in terms of BLEU scores for both the NMT (+8.05) and MMT (+11.03) systems.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

新闻领域英语-印地语低资源多模态神经机器翻译

在机器翻译系统中引入多种输入模态是机器翻译研究人员越来越关注的问题。与多模态机器翻译(MMT)任务的公开可用数据集不同，其中的标题是简短的图像描述，新闻标题提供了对图像内容的更详细的描述。结果，发现了许多与特定人员、地点等有关的已命名实体。在本文中，我们获取了英语和印地语的两个单语新闻数据集，并将其与图像配对，以生成一个合成的英语-印地语平行语料库。将图像特征与对应的并行语料库相结合，利用并行语料库训练英北神经机器翻译(NMT)和英北MMT系统。我们还进行了一个系统的分析，通过1)更多的合成数据和2)通过添加回译数据来评估英语-印地语MT系统。我们的研究结果显示，NMT(+8.05)和MMT(+11.03)系统的BLEU得分均有改善。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Proceedings of the FirstWorkshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021)

自引率

0.00%

发文量