Emotion Recognition from Speech

In the article we evaluate the importance of different HMM states in an HMM-based feature extraction method used to model paralinguistic information. Specifically, we evaluate the distribution of the paralinguistic information across different states of the HMM in two different classification tasks: emotion recognition and alcoholization detection. In the task of recognizing emotions we found that the majority of emotion-related information is incorporated in the first and third state of a 3-state HMM. Surprisingly, in the alcoholization detection task we observed a somewhat equal distribution of task-specific information across all three states, resulting in constantly producing better results if more states are utilized.

Recognizing emotions is automatically and subconsciously performed by humans. It is a vital process for human-to
human communication, and thus, to achieve better human machine interaction, emotions need to be considered. Emotional
speech recognition importance is growing in several domains. Researchers have raised the impact of emotion in
multidisciplinary applications. Predicting human emotions is catching the attention of many research areas, which demand
accurate predictions in uncontrolled scenarios. Psychologists have widely studied the influence of emotional factors, on
decision-making. As example, pilot’s decision in a flight context may jeopardize several humans life . The importance of
recognizing emotion for needs of the real world use is becoming unavoidable. In real world applications, speech signals are
often corrupted by acoustic background noise. In these applications, speech enhancement is a necessary module for the emotion
recognition system. Despite recent advances in the field of automatic speech emotion recognition, recognize emotions from vocal
channel in noisy environment remains an open research problem. We proposed a system that will do Speech detection and
Continuous recognition of emotions from speech.