REVIEW: SECURE AND RELIABLE VOICE RECOGNITION SYSTEM USING DEEP NEURAL NETWORKS ACOUSTIC MODEL

In recent days, voice recognition system is used in various impending segments like banking, communication, industrial and government sectors. Most applications are used in real time which are vulnerable to various security threats. It causes big contests on various levels in data protection to tackle security threats in cyber space. Performance of secure voice recognition system is measured. Accuracy is evaluated with word Error Rate (WER) and speed is restrained with factors such as vocabulary size, language constraints, confusability speech and image recognition will be recorded at each stage processing will be monitored. Confrontation of these types of known or unknown risks, several privacy-preserving approaches have been proposed and used for either Secure Multi-party Computation (SMC) or homomorphic