Business Meeting Summarization Using Natural Language Processing(NLP)

Abstract

— Text summarization aims to condense a source text into a shorter version. Automatic data summarization is part of data mining. In order to build a corpus for this task, it is necessary to obtain the transcription of each meeting, and then to segment and align it with the corresponding manual report to produce training examples suitable for training. In this work, an MMS method combining the techniques of natural language processing (NLP), speech processing, computer vision and advanced encryption standard (AES) encryption is used to explore the rich information contained in multi-modal data, to improve the quality and security as well the key idea is to bridge and lessen the semantic gaps between multi-modal data. For audio, speech transcriptions are used. Finally, all the multi-modal aspects are considered to generate a textual summary by maximizing the salience, readability, non-redundancy. The summary so generated by text, audio is encrypted using AES encryption method (to make it secure) and sent to members of meeting on mail, from there the user can retrieve it whenever required by providing the decryption key.