IsharaKotha: A Comprehensive Avatar-based Bangla Sign Language Corpus
MD. Ashikul Islam, Prato Dewan, Md Fuadul Islam, Md. Ataullha, M. Shahidur Rahman
TL;DR
This work addresses the lack of a Bangla Sign Language corpus suitable for text-to-sign translation by introducing IsharaKotha, the first HamNoSys-based BDSL corpus comprising 3823 signs. It combines a Seq2Seq lemmatizer to derive root words with HamNoSys-to-SiGML encoding and real-time avatar animation, enabling sentence-level sign generation. A web-based evaluation interface with interpreters and a user provides a quantitative assessment, yielding an average rating of 3.14/4, demonstrating strong sign generation quality and practical potential. The corpus and pipeline lay the groundwork for a dynamic Bangla text-to-sign translation system, with future work targeting full vocabulary coverage and integration with speech recognition to enhance accessibility.
Abstract
Sign language is a vital communication medium for the hearing-impaired community, enabling effective interaction and self-expression. To help bridge the communication gap between hearing and hearing-impaired individuals, a text-to-sign translation system is essential. Such systems can also support learners interested in acquiring sign language skills. This work presents IsharaKotha, the first HamNoSys-based Bangla Sign Language corpus, containing 3823 words. A deep learning based lemmatizer was integrated to extract root words, enabling sign generation for complete sentences. An evaluation interface was developed to assess the quality of sign animations for letters, digits, and sentences. Two professional interpreters and one real sign language user rated the animations using categorical numeric scores. The system achieved an average rating of 3.14 out of 4.00, indicating high quality performance between Good and Excellent. These results demonstrate the potential of IsharaKotha to support future advancements in dynamic sign language translation systems. The evaluation system is available at http://bdsl-isharakotha.ap-1.evennode.com