CWordle: A visual analytics system for extracting the topics of speech

Abstract

Speech information is a large part of the Internet data type, which represents many valuable resources in the real world. However, extracting topics of speech is much challenging because there is not only a requirement for an analyzation of multi-level speech data, but also a requirement for valid methods for filtering and displaying extraction results. Meanwhile, the time consumption is also an important factor. Our work provides a visualization system for analyzation and extraction of speech. By means of case studies of real-world speech data, the practicability and effectiveness of the method are demonstrated.