Handwriting and Speech‐Based Secured Multimodal Biometrics Identification Technique
Swathi Gowroju, Vudata Ramya Swathi, Ankita Tiwari · 2023
Biometric systems are crucial for security in various industries, especially banking and law enforcement. While unimodal biometric systems have been widely studied, multimodal biometrics is emerging as a critical pattern recognition component. This proposed work focuses on developing a secure authentication procedure that uses voice and signature recognition in a multimodal system for higher accuracy and lower error rates. The proposed remedy is assessed using the Kaggle TensorFlow Speech Recognition Challenge dataset. Our findings and discussions demonstrate that the suggested approach can achieve an accuracy rate of approximately 96.05%, meeting our goal, and low FAR and FRR, enhancing the multimodal authenticity of our system. This study contributes to developing robust and reliable multimodal biometric systems with significant implications for various security applications.