Recognizing and supporting children’s emotional growth can be challenging, often leading to upsets in their emotional regulation since they lack an internal alarm system to regulate their emotions. It is important to observe their continuous mood swings and emotions to assess their psychological state. In general, younger people are faster in expressing their emotions rather than elders. Emotions that are naturally exhibited if suppressed would lead to psychological traumas. Also, emotional deregulation has a detrimental effect on brain function and interpersonal abilities may likely to lead personality disorder issues. Young adults can address the impacts of childhood trauma and lead fulfilling lives with the help of efficient, evidence-based treatment or by any means of smart emotion recognition alert systems. This paper proposes a smart emotion recognition alert system based on real emotion features that will be collected in real-time. The major concerns of emotions are sadness, fear, crying, screaming, shouting, anger and frustration which are collected from the supported sensors built around the design, and will get processed by ML algorithms to get back with voice command notes for regulation. The proposed model is tested using a dataset with emotions captured by children ages 4 to elders 50. The clustering algorithm applied in assessing the accuracy along with voice modules and face detection support, observed the accuracy level reached a maximum of 72.5%. The emotion detection system alone will not be adequate in this regard as the outcomes demonstrated that, to return to normal while lessening the strength of the emotion, a voice alarm system is also necessary.