The purpose of extractive summarization is to automatically select a number of indicative sentences, passages, or paragraphs from the original document according to a target summarization ratio and then sequence them to form a concise summary. In this paper, we proposed the use of probabilistic latent topical information for extractive summarization of spoken documents. Various kinds of modeling structures and learning approaches were extensively investigated. In addition, the summarization capabilities were verified by comparison with several conventional spoken document summarization models. The experiments were performed on the Chinese broadcast news collected in Taiwan. Noticeable performance gains were obtained. The proposed summarization technique has also been properly integrated into our prototype system for voice retrieval of Mandarin broadcast news via mobile devices.
ASJC Scopus subject areas
- Signal Processing
- Computer Vision and Pattern Recognition
- Artificial Intelligence