Measuring Customer Satisfaction through Speech Using Valence-Arousal Approach

There had been many empirical researched demonstrating the important link between customer satisfaction and sales performance, as such many Customer Satisfaction index (CSI) were developed. Almost all CSI to date uses the survey or questionnaire method, which has its flaws. In order to quantify the CSI, we propose the use of speech analysis based on the affective space model where the valence and arousal of the customer can be extracted, indicating their immediate emotion. Speech were recorded and relevant features extracted using Mel Frequency Cepstral Coefficient (MFCC) coupled with Adaptive Neuro Fuzzy Inference System (ANFIS) with subtractive clustering for classification. The network were trained to responds to the two dimensional Affective Space Model (ASM) classification which consist of valence and arousal. Further analysis were carried out to understand the impact of neutral on the accuracy of the classification. Experimental results show that recognition rate for measuring satisfaction is 40% and neutral emotion obtained the highest recognition with 58%. Such analysis can help in understanding the satisfaction and dissatisfaction of customers based on speech with improving the accuracy.