Social Signal Processing for Evaluating Conversations Using Emotion Analysis and Sentiment Detection

摘要: Social Signal Processing is vital in discovering the intensity of sentiments and emotions expressed by individuals based on their facial expressions gestures, phrases tone usage. Continuous monitoring customer satisfaction an utmost important aspect industries such as banking finance. Detecting dissatisfied customers among large data captured during conversations a challenging task. In this paper, we present multi-modality framework to analyze levels, especially determining customers, using image, speech text analysis. We employ two-level synthesis: (a) emotion analysis signals sentiment detection from data, which converted same speech, (b) image data. Our approach helps better understanding identification exact perspective with respect services provided industries.