Gender Recognition Using ANN and Forward Rajan Transform Inclusive of Transgender Identity

Low-resource languages are at risk of extinction due to factors like globalization and the dominance of widely spoken languages. Creating a comprehensive low-resource language dataset for gender recognition from marginalized communities is imperative to ensure fairness, equality, and unbiased representation in gender recognition technology. It also aids equitable representation of gender identities and linguistic diversity. The proposed work addresses the low-resource language dataset collected from the transgender community. This work comprises dataset creation of three classes such as male, female, and transgender, pre-processing, Forward Rajan Transform (FRT) feature extraction, and ANN classification for gender recognition. The performance of the proposed model compared with other ANN models and the proposed model provided a superior accuracy of 96.4%. The experimental results show the efficacy of the proposed model compared with SOTA approaches.