@inproceedings{dabre-song-2024-nicts,
title = "{NICT}{'}s Cascaded and End-To-End Speech Translation Systems using Whisper and {I}ndic{T}rans2 for the {I}ndic Task",
author = "Dabre, Raj and
Song, Haiyue",
editor = "Salesky, Elizabeth and
Federico, Marcello and
Carpuat, Marine",
booktitle = "Proceedings of the 21st International Conference on Spoken Language Translation (IWSLT 2024)",
month = aug,
year = "2024",
address = "Bangkok, Thailand (in-person and online)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.iwslt-1.3",
doi = "10.18653/v1/2024.iwslt-1.3",
pages = "17--22",
abstract = "This paper presents the NICT{'}s submission for the IWSLT 2024 Indic track, focusing on three speech-to-text (ST) translation directions: English to Hindi, Bengali, and Tamil. We aim to enhance translation quality in this low-resource scenario by integrating state-of-the-art pre-trained automated speech recognition (ASR) and text-to-text machine translation (MT) models. Our cascade system incorporates a Whisper model fine-tuned for ASR and an IndicTrans2 model fine-tuned for MT. Additionally, we propose an end-to-end system that combines a Whisper model for speech-to-text conversion with knowledge distilled from an IndicTrans2 MT model. We first fine-tune the IndicTrans2 model to generate pseudo data in Indic languages. This pseudo data, along with the original English speech data, is then used to fine-tune the Whisper model. Experimental results show that the cascaded system achieved a BLEU score of 51.0, outperforming the end-to-end model, which scored 19.1 BLEU. Moreover, the analysis indicates that applying knowledge distillation from the IndicTrans2 model to the end-to-end ST model improves the translation quality by about 0.7 BLEU.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dabre-song-2024-nicts">
<titleInfo>
<title>NICT’s Cascaded and End-To-End Speech Translation Systems using Whisper and IndicTrans2 for the Indic Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Raj</namePart>
<namePart type="family">Dabre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haiyue</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 21st International Conference on Spoken Language Translation (IWSLT 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Elizabeth</namePart>
<namePart type="family">Salesky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcello</namePart>
<namePart type="family">Federico</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marine</namePart>
<namePart type="family">Carpuat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand (in-person and online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper presents the NICT’s submission for the IWSLT 2024 Indic track, focusing on three speech-to-text (ST) translation directions: English to Hindi, Bengali, and Tamil. We aim to enhance translation quality in this low-resource scenario by integrating state-of-the-art pre-trained automated speech recognition (ASR) and text-to-text machine translation (MT) models. Our cascade system incorporates a Whisper model fine-tuned for ASR and an IndicTrans2 model fine-tuned for MT. Additionally, we propose an end-to-end system that combines a Whisper model for speech-to-text conversion with knowledge distilled from an IndicTrans2 MT model. We first fine-tune the IndicTrans2 model to generate pseudo data in Indic languages. This pseudo data, along with the original English speech data, is then used to fine-tune the Whisper model. Experimental results show that the cascaded system achieved a BLEU score of 51.0, outperforming the end-to-end model, which scored 19.1 BLEU. Moreover, the analysis indicates that applying knowledge distillation from the IndicTrans2 model to the end-to-end ST model improves the translation quality by about 0.7 BLEU.</abstract>
<identifier type="citekey">dabre-song-2024-nicts</identifier>
<identifier type="doi">10.18653/v1/2024.iwslt-1.3</identifier>
<location>
<url>https://aclanthology.org/2024.iwslt-1.3</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>17</start>
<end>22</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T NICT’s Cascaded and End-To-End Speech Translation Systems using Whisper and IndicTrans2 for the Indic Task
%A Dabre, Raj
%A Song, Haiyue
%Y Salesky, Elizabeth
%Y Federico, Marcello
%Y Carpuat, Marine
%S Proceedings of the 21st International Conference on Spoken Language Translation (IWSLT 2024)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand (in-person and online)
%F dabre-song-2024-nicts
%X This paper presents the NICT’s submission for the IWSLT 2024 Indic track, focusing on three speech-to-text (ST) translation directions: English to Hindi, Bengali, and Tamil. We aim to enhance translation quality in this low-resource scenario by integrating state-of-the-art pre-trained automated speech recognition (ASR) and text-to-text machine translation (MT) models. Our cascade system incorporates a Whisper model fine-tuned for ASR and an IndicTrans2 model fine-tuned for MT. Additionally, we propose an end-to-end system that combines a Whisper model for speech-to-text conversion with knowledge distilled from an IndicTrans2 MT model. We first fine-tune the IndicTrans2 model to generate pseudo data in Indic languages. This pseudo data, along with the original English speech data, is then used to fine-tune the Whisper model. Experimental results show that the cascaded system achieved a BLEU score of 51.0, outperforming the end-to-end model, which scored 19.1 BLEU. Moreover, the analysis indicates that applying knowledge distillation from the IndicTrans2 model to the end-to-end ST model improves the translation quality by about 0.7 BLEU.
%R 10.18653/v1/2024.iwslt-1.3
%U https://aclanthology.org/2024.iwslt-1.3
%U https://doi.org/10.18653/v1/2024.iwslt-1.3
%P 17-22
Markdown (Informal)
[NICT’s Cascaded and End-To-End Speech Translation Systems using Whisper and IndicTrans2 for the Indic Task](https://aclanthology.org/2024.iwslt-1.3) (Dabre & Song, IWSLT 2024)
ACL