Skip to content
conftrace_
2024 ACL ACL 2024

Empowering Low-Resource Language Translation: Methodologies for Bhojpuri-Hindi and Marathi-Hindi ASR and MT

Abstract

AbstractThe paper describes our submission for the unconstrained track of ‘Dialectal and Low-Resource Task’ proposed in IWSLT-2024. We designed cascaded Speech Translation systems for the language pairs Marathi-Hindi and Bhojpuri-Hindi utilising and fine-tuning different pre-trained models for carrying out Automatic Speech Recognition (ASR) and Machine Translation (MT).

🌉 Interdisciplinary Bridge - Natural Language Processing and Speech & Audio

Authors

Harpreet Singh Anand, Amulya Ratna Dash, Yashvardhan Sharma

Topics

Natural Language Processing > Applications > Machine Translation Speech & Audio > Recognition > Automatic Speech Recognition

Keywords

machine translation automatic speech recognition low-resource language pre-trained model cascaded system
Download PDF

More from ACL 2024

All ACL 2024 papers →
ALADAN at IWSLT24 Low-resource Arabic Dialectal Speech Translation Task
Enhancing Translation Accuracy of Large Language Models through Continual Pre-Training on Parallel Data
The KIT Speech Translation Systems for IWSLT 2024 Dialectal and Low-resource Track
Recent Highlights in Multilingual and Multimodal Speech Translation
Word Order in English-Japanese Simultaneous Interpretation: Analyses and Evaluation using Chunk-wise Monotonic Translation
Leveraging Synthetic Audio Data for End-to-End Low-Resource Speech Translation