Intelligent telecommunication system using semantic-based information retrieval

Abstract

Artificial intelligence refers to the ability by which a system studies human ideas and applies them in computerized machines. One of the advantages of a speech recognition system is to simultaneously convert the user’s voice into text. The current telephony system provides a feature of recording a voice call, but it has no processing capability to analyze a phone call. In our project, we introduce an intelligent agent into the telephony system. An unattended phone call is recorded and a recognizer converts the analog signal into a digital signal for speech processing. The analyzer consists of a database which holds some important preprocessed keywords and the intelligent agent can formulate the state of the situation. In case of urgency, the message is examined by the parser to identify the time of the incident and the sentiment analyzer has been used to avoid prank calls. Thus, an epistle would be sent to the user in case of exigency or the voice message is stored by default method.