In the last few years, a lot of emphasis has been given to enhancing agricultural production as well as improving land productivity to cope with the increasing food demand. However, post-harvest losses are a serious concern for food grain loss due to improper storage, insects, rodents, and microorganisms. The main aim of this project is to develop a speech recognition model for the early detection of pests by analyzing the sounds of the insects at different stages of the stored food grains. In the second stage, a vibration separator-based approach will be used to protect the stored food grains from pest infestation from the beginning.