Enhancing call center operations through semantic voice analysis

Abstract

A call center is a central place where customer care agents handle queries of customers over the telephone to fulfill and satisfy their needs. Information searching delay is an increasingly and important problem in a call center environment because of multiple and not similar subsystems in operation. Developed application is running stand-alone and no need to install, and it uses.NET framework and NAudio open-source library for enhancing signals and converting enhanced voice signal into text. Application is consisted with a prediction algorithm and a categorization functionality which is capable for correcting falsely recognized text and eliminates the need of following highly time and human memory consuming file menus or hyperlinks. This will increase the accuracy of the recognized sentences, lead to better information searching functionality, and propose the most probable locations of the required information. Application was tested with several users and achieved 90% accuracy for information categorization.