Abstract
AbstractThe intensive research of speech emotion recognition introduced a huge collection of speech emotion features. Large feature sets complicate the speech emotion recognition task. Among various feature selection and transformation techniques for one-stage classification, multiple classifier systems were proposed. The main idea of multiple classifiers is to arrange the emotion classification process in stages. Besides parallel and serial cases, the hierarchical arrangement of multi-stage classification is most widely used for speech emotion recognition. In this paper, we present a sequential-forward-feature-selection-based multi-stage classification scheme. The Sequential Forward Selection (SFS) and Sequential Floating Forward Selection (SFFS) techniques were employed for every stage of the multi-stage classification scheme. Experimental testing of the proposed scheme was performed using the German and Lithuanian emotional speech datasets. Sequential-feature-selection-based multi-stage classification outperformed the single-stage scheme by 12–42 % for different emotion sets. The multi-stage scheme has shown higher robustness to the growth of emotion set. The decrease in recognition rate with the increase in emotion set for multi-stage scheme was lower by 10–20 % in comparison with the single-stage case. Differences in SFS and SFFS employment for feature selection were negligible.
Highlights
Human–computer interaction has become an everyday routine nowadays
Large feature sets complicate the speech emotion recognition task as the number of analyzed speech patterns becomes lower than the feature order
Sequential Forward Selection and Sequential Floating Forward Selection techniques were applied for multi-stage classification of speech emotions
Summary
Human–computer interaction has become an everyday routine nowadays. The knowledge of computer science, sociology psychology, data visualization, and other fields are integrated to improve this interaction. Implementing speech communication into interaction process brings some challenges. Besides obvious tasks like noise removal problem, speech recognition or speaker identification, the question of speaker’s emotional state arises. The emotional state of the speaker affects his speech inevitably, making aforementioned tasks more complicated. A well-timed and accurate identification of particular emotional states of the client would help to optimize the work process of customer service centers and call-centers by redirecting the calling person to agents with appropriate qualification [1]. Speech emotion identification could be integrated into personal assistance systems helping us to drive car, staying at home, hospitals or shopping. Speech emotion recognition is a common classification task with three major steps: the feature set formation, the training process of the classifier, and the process of decision-making about an unknown emotional pattern. The language-independent or language-specific emotion features, feature selection and feature sets, classification scheme, language effect on emotion recognition and other questions need to be answered to achieve a robust and reliable speech emotion recognition
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have