@inproceedings{indurthi-etal-2023-clad,
title = "{CLAD}-{ST}: Contrastive Learning with Adversarial Data for Robust Speech Translation",
author = "Indurthi, Sathish and
Chollampatt, Shamil and
Agrawal, Ravi and
Turchi, Marco",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.emnlp-main.560",
doi = "10.18653/v1/2023.emnlp-main.560",
pages = "9049--9056",
abstract = "The cascaded approach continues to be the most popular choice for speech translation (ST). This approach consists of an automatic speech recognition (ASR) model and a machine translation (MT) model that are used in a pipeline to translate speech in one language to text in another language. MT models are often trained on the well-formed text and therefore lack robustness while translating noisy ASR outputs in the cascaded approach, degrading the overall translation quality significantly. We address this robustness problem in downstream MT models by forcing the MT encoder to bring the representations of a noisy input closer to its clean version in the semantic space. This is achieved by introducing a contrastive learning method that leverages adversarial examples in the form of ASR outputs paired with their corresponding human transcripts to optimize the network parameters. In addition, a curriculum learning strategy is then used to stabilize the training by alternating the standard MT log-likelihood loss and the contrastive losses. Our approach achieves significant gains of up to 3 BLEU scores in English-German and English-French speech translation without hurting the translation quality on clean text.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="indurthi-etal-2023-clad">
<titleInfo>
<title>CLAD-ST: Contrastive Learning with Adversarial Data for Robust Speech Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sathish</namePart>
<namePart type="family">Indurthi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shamil</namePart>
<namePart type="family">Chollampatt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ravi</namePart>
<namePart type="family">Agrawal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Turchi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="family">Pino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The cascaded approach continues to be the most popular choice for speech translation (ST). This approach consists of an automatic speech recognition (ASR) model and a machine translation (MT) model that are used in a pipeline to translate speech in one language to text in another language. MT models are often trained on the well-formed text and therefore lack robustness while translating noisy ASR outputs in the cascaded approach, degrading the overall translation quality significantly. We address this robustness problem in downstream MT models by forcing the MT encoder to bring the representations of a noisy input closer to its clean version in the semantic space. This is achieved by introducing a contrastive learning method that leverages adversarial examples in the form of ASR outputs paired with their corresponding human transcripts to optimize the network parameters. In addition, a curriculum learning strategy is then used to stabilize the training by alternating the standard MT log-likelihood loss and the contrastive losses. Our approach achieves significant gains of up to 3 BLEU scores in English-German and English-French speech translation without hurting the translation quality on clean text.</abstract>
<identifier type="citekey">indurthi-etal-2023-clad</identifier>
<identifier type="doi">10.18653/v1/2023.emnlp-main.560</identifier>
<location>
<url>https://aclanthology.org/2023.emnlp-main.560</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>9049</start>
<end>9056</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CLAD-ST: Contrastive Learning with Adversarial Data for Robust Speech Translation
%A Indurthi, Sathish
%A Chollampatt, Shamil
%A Agrawal, Ravi
%A Turchi, Marco
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F indurthi-etal-2023-clad
%X The cascaded approach continues to be the most popular choice for speech translation (ST). This approach consists of an automatic speech recognition (ASR) model and a machine translation (MT) model that are used in a pipeline to translate speech in one language to text in another language. MT models are often trained on the well-formed text and therefore lack robustness while translating noisy ASR outputs in the cascaded approach, degrading the overall translation quality significantly. We address this robustness problem in downstream MT models by forcing the MT encoder to bring the representations of a noisy input closer to its clean version in the semantic space. This is achieved by introducing a contrastive learning method that leverages adversarial examples in the form of ASR outputs paired with their corresponding human transcripts to optimize the network parameters. In addition, a curriculum learning strategy is then used to stabilize the training by alternating the standard MT log-likelihood loss and the contrastive losses. Our approach achieves significant gains of up to 3 BLEU scores in English-German and English-French speech translation without hurting the translation quality on clean text.
%R 10.18653/v1/2023.emnlp-main.560
%U https://aclanthology.org/2023.emnlp-main.560
%U https://doi.org/10.18653/v1/2023.emnlp-main.560
%P 9049-9056
Markdown (Informal)
[CLAD-ST: Contrastive Learning with Adversarial Data for Robust Speech Translation](https://aclanthology.org/2023.emnlp-main.560) (Indurthi et al., EMNLP 2023)
ACL