Low Resource Neural Machine Translation: A Benchmark for Five African Languages – ICLR 2020

Low Resource Neural Machine Translation: A Benchmark for Five African Languages – ICLR 2020

Published on July 25, 2020

There are over 7000 languages and language varieties in the world. Despite the progress in Neural Machine Translation, there still shortcomings if real low-resource languages are considered.

In this talk, Surafel Melaku Lakew talks about the advancements in low-resource language translation tasks while looking at 5 languages. The languages are Swahili, Somali, Amharic, Tigrigna and Oromo

Learn more about this project: Low Resource Neural Machine Translation: A Benchmark for Five African Languages by Surafel Melaku Lakew · Apr 26, 2020 · ICLR 2020

This talk was delivered in Addis Ababa, Ethiopia at the ICLR 2020 “AfricaNLP Workshop – Putting Africa on the NLP Map” via the AI4D Africa program with support from IDRC, Sida and Knowledge 4 All Foundation