Skip to content
conftrace_
2025 ACL ACL 2025

CDAC-SVNIT submission for IWSLT 2025 Indic track shared task

Abstract

AbstractIn this paper, we designed a Speech-to-Text Translation (ST) system to translate English into Hindi, Bengali, and Tamil, and vice versa. We explored both cascaded and End-to-End (E2E) approaches as part of the IWSLT 2025 Indic shared task.

Authors

Mukund K. Roy, Karunesh Arora, Praveen Kumar Chandaliya, Rohit Kumar, Pruthwik Mishra

Topics

Machine Learning > Application Areas > Domain Adaptation

Keywords

machine translation end-to-end translation speech translation speech-to-text translation cascaded system indic language cascaded translation
Download PDF

More from ACL 2025

All ACL 2025 papers →
FFSTC 2: Extending the Fongbe to French Speech Translation Corpus
HENT-SRT: Hierarchical Efficient Neural Transducer with Self-Distillation for Joint Speech Recognition and Translation
Swiss German Speech Translation and the Curse of Multidialectality
NAVER LABS Europe Submission to the Instruction-following Track
JU-CSE-NLP’s Cascaded Speech to Text Translation Systems for IWSLT 2025 in Indic Track
NYA’s Offline Speech Translation System for IWSLT 2025