Voice recognition using machine learning and central database to enhance security system

Abstract

Voice recognition refers to the purpose of interpreting voice or identifying any individual voice. In modern days of technological advancement voice recognition has been playing an integral part behind many machine learning algorithms. Furthermore, speech recognition, alternatively referred to as voice recognition, can help us immensely in particular scenarios such as in building better access control system and security system. Voice detection and comparison is a challenging problem because the traditional methods of speech recognition are not on par with human capabilities. In modern machine learning methodologies there is a vast potential to overcome barriers of detecting human speech. The voice is a simple medium people use for everyday communication, so it can be used to improve security system by utilizing voice recognition identifying an individual. This article focuses on enhancing security system by deep learning based approach of voice recognition. Moreover, the article further elaborates about using available datasets from a central database which is used for voice detection and comparison. The focal point of this article is to apply the most suitable methodologies of machine learning and deep learning to detect any individual by the prosodic feature of speech from a given central database.