The extent, to which people listen to and perceive the speech content at different noise levels varies from individual to individual. In past research projects, the speech intelligibility was determined by rating assessment, which suffered from variation of subjects' physical features. The purpose of this study is to investigate electroencephalography (EEG) by implementing multi-variate Temporal Response Function (mTRF) to examine neural responses to speech stimuli at different sound and noise levels. The result of this study shows that the front-central area of the brain clearly shows the envelope entrainment to speech stimuli. ©2019 IEEE.