Human-machine communication can be substantially enhanced by the inclusion of high-quality real-time recognition of spontaneous emotional expressions. However, successful recognition of such expressions can be negatively impacted by factors such as sudden variations of lighting, or intentional obfuscation. In addition, the presentation and meaning of emotional expressions can vary significantly, depending on external factors such as the environment within which the emotions are expressed, and the culture of the expressor. As an example, an emotion recognition model trained on a regionally-specific database, collected from North America, might fail to recognize standard emotional expressions from another region, such as East Asia. To address the problem of regional and cultural bias in emotion recognition from facial expressions, we propose a meta-model that fuses multiple emotional cues and features. The proposed approach integrates macro-expression, micro-expression, action level features and image features into a multi-cues emotion model (MCAM). The results show that the proposed meta-classifier (MCAM) reduces the bias factors and achieves successful emotion recognition in most of the evaluated scenarios. Additionally, we also identified certain cues and features of the data-sets that preclude the design of the perfect unbiased classifier.