This repository contains the dataset and artifacts associated with the paper Bandwidth-Efficient and Privacy-Preserving Edge-Cloud Many-to-Many Speech Translation.
Edge-cloud Speech Recognition and Translation (ESRT) is a privacy-preserving and bandwidth-efficient collaborative edge-cloud MLLM framework. It addresses English-centric biases in many-to-many translation scaling across 45 languages ($45 \times 44$ directions).… See the full description on the dataset page:
https://huggingface.co/datasets/yxdu/fleurs_eng_test.
