Dysarthria Speech Disorder Assessment Using Genetic Algorithm (GA)-Based Layered Recurrent Neural Network

A speech issue known as dysarthria occurs because of muscular weakness and nerve damage following a stroke, an infection in the brain, or a brain injury. Many Speech Therapies are involved in assisting people with Dysarthria Speech Disorder. Dysarthria is also said to have an influence on the comprehensive potentiality, speech accessibility and a specific persons’ capability to unite and interconnect in day to day chores. The Purpose of this work is for the early detection of speech disorder hence would have a positive influence on the quality of life. An effective new approach for Dysarthria speech recognition requires optimization and learning patterns to get better accuracy measurements. In this, Cuckoo Search Optimization Technique is used for better accuracy results. Genetic Algorithm (GA)-based Layered Recurrent Neural Network Improved Cuckoo Search Optimization (GALRNN-ICSO) method is used for Dysarthria Speech Recognition. The suggested GALRNN-ICSO method's primary goal is to improve test-accuracy measurement via accuracy rate as well as the precision and RMSE rate while recognizing Dysarthria Speech. The objective is to increase the accuracy and cuts down the time required for Dysarthria Speech recognition. From the experimental result, proposed GALRNN-ICSO method ensures more accuracy with precise assessment of disorder compared to existing state-of-the-art methods.