Atypical or delayed emotion processing has been repeatedly reported in individuals with Autism Spectrum Disorder (ASD); the overwhelming research focus has been on examining the level of their impairments to perceive emotions via (mostly static) facial expressions and body movement. While it is difficult for them to read others' emotions (including those with ASD); it might be just as challenging for neuro-typical (NT) individuals to recognize their emotions. Instead of pursuing a deeper understanding down the path on the expressive abilities of individuals with ASD, in our on-going study, we focus on the integration of a naturalistic multi-sensory environment (including a facial expression recognition module through Kinect V2's HD Face API) to help NT individuals "read" the emotions of children with ASD. In this short report, we focus on our system design and offered some early insights derived from initial testing on the potential of such environment.