2M-BELEBELE: Highly Multilingual Speech and American Sign Language Comprehension Dataset
This work addresses the need for multilingual and multimodal comprehension datasets for researchers in natural language processing and accessibility, though it is incremental as it extends an existing dataset.
The authors tackled the problem of multilingual speech and sign language comprehension by creating the 2M-BELEBELE dataset, which covers 74 spoken languages and American Sign Language, and found that speech comprehension accuracy is about 2-3% lower on average compared to reading comprehension in evaluations.
We introduce the first highly multilingual speech and American Sign Language (ASL) comprehension dataset by extending BELEBELE. Our dataset covers 74 spoken languages at the intersection of BELEBELE and FLEURS, and one sign language (ASL). We evaluate 2M-BELEBELE dataset for both 5-shot and zero-shot settings and across languages, the speech comprehension accuracy is ~ 2-3% average lower compared to reading comprehension.