Children with High-Functioning Autism (HFA) have difficulty understanding and using emotional expressions. A variety of smart media, such as tablet PCs and robots, are being used to train children with autism spectrum disorders on emotional vocabulary use through language therapy. In this work, we explore the effectiveness of robot-based intervention (compared with PC-based intervention) on children with HFA, in particular, we evaluate the impact of using emotional stories to understand how intervention impacts the total number (and the diversity) of words to express their emotions. Our initial results revealed that children with HFA showed significantly higher performance when using robot intervention than PCs, in terms of the total number of emotional words used. Our study thus suggests that robot-based intervention can be used as a potential therapeutic tool for training children with HFA to increase their ability to use words to express their emotions.