Emotions and their intimate relationship with how humans interact with each other are currently studied as an alternative to stimulate what is called as Human-Robot Interaction (HRI). Different strategies that use emotions in robots as a central element have been developed to fulfil this goal, which has allowed the improvement of the interaction quality between a robot and a human. In this paper, different models and APIs are discussed and integrated for the processing of visual and audio stimulus in order to deliver the evoked emotions to a deterministic system and finally get the intelligent execution of behaviors in a NAO robot. The tests carried out with different case studies validated that the system responds according to the characteristics of each audiovisual stimulus and this does not have a bias towards a particular case.