Pytorch based BERT, mBART and NMT training
Multilingual Pre-training with Language and Task Adaptation for Multilingual Text Style Transfer (ACL 2022)
This project is about a neural machine translation in wolof language to french .
Repository for distillation of English, Chinese, Korean Multilingual BART
Two New Datasets for Italian-Language Abstractive Text Summarization
Translate Arabic documents to English effortlessly with this Streamlit web application. The app supports PDF, text (TXT), and Word document (DOCX) formats. Simply upload your file, and the app will display the original text alongside its English translation. Powered by the MBart model from Hugging Face Transformers.
A two-stage pipeline for generating Hindi captions from video using BLIP and fine-tuned mBART-50, without requiring any Hindi video-caption datasets.
Multilingual Speech Translator with Whisper + mBART-50 + gTTS
A powerful AI-powered application that provides audio transcription, text summarization, and multi-language translation capabilities. Built with FastAPI and deployed on Hugging Face Spaces.
Multilingual translation system based on mBART, using Spanish as a pivot language for German, Dutch, and Hindi.
A project on controlled Russian text simplification.
Collaborative filtering-based anime recommendation system.
NLP Assignment 2
Continuous Sign Language Recognition & Translation (video → gloss → text) on PHOENIX-2014T: 3-stream CTC (RGB Swin3D, Hypergraph Skeleton, Cross-Attention) + Weighted ROVER + mBART-50.
Multi-language summarization with mBART, TextRank, and evaluation
Kazakh text summarization research — mono and cross-lingual (kk↔ru)
A transliteration system that converts English text to Hindi, Bengali, and Tamil using fine-tuned mBART-50 with LoRA optimization and CTranslate2 acceleration.
To associate your repository with the mbart topic, visit your repo's landing page and select "manage topics."