Prosodic feature based speech emotion recognition at segmental and supra segmental levels

STEP: Spatial Temporal Graph Convolutional Networks for Emotion Perception from Gaits

“…Prior methods that use faces as input commonly track action units on the face such as points on the eyebrow, cheeks and lips (Fabian Benitez-Quiroz, Srinivasan, and Martinez 2016), or track eye movements (Schurgin et al 2014) and facial expressions (Majumder, Behera, and Subramanian 2014). Speechbased emotion perception methods use either spectral features or prosodic features like loudness of voice, difference in tones and changes in pitch (Jacob and Mythili 2015).…”

“…Prior methods that use faces as input commonly track action units on the face such as points on the eyebrow, cheeks and lips [16], or track eye movements [39] and facial expressions [33]. Speech-based emotion perception methods use either spectral features or prosodic features like loudness of voice, difference in tones and changes in pitch [24]. With the rising popularity of deep learning, there is considerable work on developing learned features for emotion detection from large-scale databases of faces [51,53] and speech signals [12].…”

“…Human emotion recognition using intelligent systems is an important socio-behavioral task that arises in various applications, including behavior prediction (Denham et al 2000), surveillance (Arunnehru and Geetha 2017), robotics (Bauer and others 2009), affective computing (Yates et al 2017;Atcheson, Sethu, and Epps 2017), etc. Current research in perceiving human emotion predominantly uses facial cues (Fan et al 2016), speech (Jacob and Mythili 2015), or physiological signals such as heartbeats and respiration rates (Zhao, Adib, and Katabi 2016). These techniques have been used to identify and classify broad emotions including happiness, sadness, anger, disgust, fear and other combinations (Ekman and Friesen 1967).…”

Learning Unseen Emotions from Gestures via Semantically-Conditioned Zero-Shot Perception with Adversarial Autoencoders

“…Emotion learning as an area of research is integral to a variety of domains, including human-computer interaction, robotics (Liu et al 2017) and affective computing (Yates et al 2017). Existing research in emotion recognition has leveraged aspects such as facial expressions (Liu et al 2017), speech (Jacob and Mythili 2015), gestures and gaits (Bhattacharya et al 2020a) to gauge an individual's emotional state. Studies in psychology indicate that humans perceive emotions by observing affective features such as arm swing rate, posture, and frequency of movements.…”

Take an Emotion Walk: Perceiving Emotions from Gaits Using Hierarchical Attention Pooling and Affective Mapping

“…Humans perceive each other's emotions and moods through verbal cues such as speech [47,27] and text [56,12], as well as through non-verbal cues or affective features [48], including eye-movements [50], facial expressions [18], tone of voice, postures [4], and walking styles [32]. Understanding these perceived emotions shapes people's interactions and experiences, especially when performing tasks in collaborative or competitive environments [6].…”