Experiences of Adapting Multimodal Machine Translation Techniques for Hindi

Baban Gain, Dibyanayan Bandyopadhyay, Asif Ekbal


Abstract
Multimodal Neural Machine Translation (MNMT) is an interesting task in natural language processing (NLP) where we use visual modalities along with a source sentence to aid the source to target translation process. Recently, there has been a lot of works in MNMT frameworks to boost the performance of standalone Machine Translation tasks. Most of the prior works in MNMT tried to perform translation between two widely known languages (e.g. English-to-German, English-to-French ). In this paper, We explore the effectiveness of different state-of-the-art MNMT methods, which use various data oriented techniques including multimodal pre-training, for low resource languages. Although the existing methods works well on high resource languages, usability of those methods on low-resource languages is unknown. In this paper, we evaluate the existing methods on Hindi and report our findings.
Anthology ID:
2021.mmtlrl-1.7
Volume:
Proceedings of the First Workshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021)
Month:
September
Year:
2021
Address:
Online (Virtual Mode)
Venue:
MMTLRL
SIG:
Publisher:
INCOMA Ltd.
Note:
Pages:
40–44
Language:
URL:
https://aclanthology.org/2021.mmtlrl-1.7
DOI:
Bibkey:
Cite (ACL):
Baban Gain, Dibyanayan Bandyopadhyay, and Asif Ekbal. 2021. Experiences of Adapting Multimodal Machine Translation Techniques for Hindi. In Proceedings of the First Workshop on Multimodal Machine Translation for Low Resource Languages (MMTLRL 2021), pages 40–44, Online (Virtual Mode). INCOMA Ltd..
Cite (Informal):
Experiences of Adapting Multimodal Machine Translation Techniques for Hindi (Gain et al., MMTLRL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/auto-file-uploads/2021.mmtlrl-1.7.pdf
Data
HindEnCorp