Code for my MSc Dissertation titled: "Robustness of Machine Translation for Low-Resource Languages."
-
Updated
Nov 22, 2021 - Shell
Code for my MSc Dissertation titled: "Robustness of Machine Translation for Low-Resource Languages."
This represents the first instance of NSEC, a spelling error correction system, based on a CNN-NMT approach. Due to the black-box behaviour this one was replaced by a modularised and observable version.
A simple implementation of Chinese-couplet generation using fairseq.
The repository contains model implementations and data described in the paper: From Dataset Recycling to Multi-Property Extraction and Beyond.
Translating English text to Persian using Fairseq-py
Massively Multilingual Speech (MMS) - Text To Speech Webview app - 1000+ languages
attention pruning
Code for the paper "Does Joint Training Really Help Cascaded Speech Translation?" (EMNLP 2022)
Автоматическое реферирование текста (TextRank, mBART)
The Referential Reader: A Recurrent Entity Network for Anaphora Resolution, published at ACL 2019
Master's Thesis in Natural Language Generation
A comprehensive list of awesome self-supervised speech representation learning papers.
Noisy machine translation, final project for 11731
Transfer Learning for Text Summarization
BioGPT is a generative language model that has been pre-trained on large amounts of biomedical literature. It is a domain-specific variant of the GPT family of language models and is designed to generate fluent descriptions for biomedical terms.
Add a description, image, and links to the fairseq topic page so that developers can more easily learn about it.
To associate your repository with the fairseq topic, visit your repo's landing page and select "manage topics."