Lip Reading Vowel-Bangla (LRV-B) Dataset

The Lip Reading Vowel-Bangla (LRV-B) Dataset is a curated collection of video recordings focused on the articulation of Bangla vowels. It is designed to support research in lip reading and visual speech recognition for the Bangla language. The dataset captures the lip movements associated with 6 (six) key Bangla vowel sounds, providing a valuable resource for developing and evaluating models in speech recognition, language processing, and assistive technologies. By concentrating on vowels, the LRV-B Dataset addresses the specific challenges of recognizing and interpreting these fundamental speech components in Bangla.