Development of an auditory emotion recognition function using psychoacoustic parameters based on the International Affective Digitized Sounds

Abstract

The purpose of this study was to develop an auditory emotion recognition function that could determine the emotion caused by sounds coming from the environment in our daily life. For this purpose, sound stimuli from the International Affective Digitized Sounds (IADS-2), a standardized database of sounds intended to evoke emotion, were selected, and four psychoacoustic parameters (i.e., loudness, sharpness, roughness, and fluctuation strength) were extracted from the sounds. Also, by using an emotion adjective scale, 140 college students were tested to measure three basic emotions (happiness, sadness, and negativity). From this discriminant analysis to predict basic emotions from the psychoacoustic parameters of sound, a discriminant function with overall discriminant accuracy of 88.9 % was produced from training data. In order to validate the discriminant function, the same four psychoacoustic parameters were extracted from 46 sound stimuli collected from another database and substituted into the discriminant function. The results showed that an overall discriminant accuracy of 63.04 % was confirmed. Our findings provide the possibility that daily-life sounds, beyond voice and music, can be used in a human-machine interface.

Negative Affect‐Associated USV Acoustic Characteristics Predict Future Excessive Alcohol Drinking and Alcohol Avoidance in Male P and NP Rats

“…The present work also suggests that we can utilize naturalistic measures, such as an analysis of emotionality (e.g., electrophysiology), to determine significant associations with abuse potential. Along these lines, a number of laboratories are actively developing tools for classifying the emotional content of speech (Choi et al., ; Cohen et al., ), but to the best of our knowledge this research has yet to be applied to alcohol abuse. Given the strong relationship between negative affect and AUD in animals and humans, it is reasonable to propose that interventions (e.g., mindfulness‐based interventions) that reduce the prevalence of negative affect might increase alcohol abstinence (i.e., prevent relapse).…”

Supervised machine learning for audio emotion recognition

“…The potential for machine learning techniques to be employed in AER, specifically using the IADS set, was previously further verified by Choi et al [12], who consider the audio clips in IADS to broadly represent those that would be encountered in daily life. Using a small set of audio features, namely loudness, sharpness, roughness and fluctuation strength, Choi and colleagues were able to demonstrate a better-than-chance discriminant function able to classify audio clips from IADS into one of three emotional factors: happiness, sadness and negativity.…”

A Preliminary Investigation on Frequency Dependant Cues for Human Emotions

“…In another research, Paul C. Vitz identified preferred tones as an inverted function of frequency and intensity [36]. Apart from that, other noteworthy researches in the field of psychoacoustics are the development of an auditory emotion recognition function using psychoacoustic parameters based on IADS [21], and emotional reactions to sounds without meaning [37].…”

“…To date, studies on emotion recognition have been predominantly based on identifying cues from facial expression data [9,17], voice and speech data [18], audio-visual data [9,10], and physiological data [9,19,20]. Among these, the area of research involving sound without verbal information as the stimulus for emotion recognition is classified as psychoacoustics [21,22]. It has the potential to be established as a distinct field of emotion recognition, as sound plays an important role in detecting cues for various primary emotions.…”