Minimum cut model for spoken lecture segmentation
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, February 2007.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2010
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/51651 |
| _version_ | 1811093445892112384 |
|---|---|
| author | Malioutov, Igor (Igor Mikhailovich) |
| author2 | Regina Barzilay. |
| author_facet | Regina Barzilay. Malioutov, Igor (Igor Mikhailovich) |
| author_sort | Malioutov, Igor (Igor Mikhailovich) |
| collection | MIT |
| description | Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, February 2007. |
| first_indexed | 2024-09-23T15:45:10Z |
| format | Thesis |
| id | mit-1721.1/51651 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T15:45:10Z |
| publishDate | 2010 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/516512019-04-12T23:35:07Z Minimum cut model for spoken lecture segmentation Malioutov, Igor (Igor Mikhailovich) Regina Barzilay. Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science. Electrical Engineering and Computer Science. Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, February 2007. Includes bibliographical references (leaves 129-132). We introduce a novel unsupervised algorithm for text segmentation. We re-conceptualize text segmentation as a graph-partitioning task aiming to optimize the normalized-cut criterion. Central to this framework is a contrastive analysis of lexical distribution that simultaneously optimizes the total similarity within each segment and dissimilarity across segments. Our experimental results show that the normalized-cut algorithm obtains performance improvements over the state-of-the-art techniques on the task of spoken lecture segmentation. Another attractive property of the algorithm is robustness to noise. The accuracy of our algorithm does not deteriorate significantly when applied to automatically recognized speech. The impact of the novel segmentation framework extends beyond the text segmentation domain. We demonstrate the power of the model by applying it to the segmentation of raw acoustic signal without intermediate speech recognition. by Igor Malioutov. S.M. 2010-02-09T16:56:45Z 2010-02-09T16:56:45Z 2006 2007 Thesis http://hdl.handle.net/1721.1/51651 500912209 eng M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission. http://dspace.mit.edu/handle/1721.1/7582 132 leaves application/pdf Massachusetts Institute of Technology |
| spellingShingle | Electrical Engineering and Computer Science. Malioutov, Igor (Igor Mikhailovich) Minimum cut model for spoken lecture segmentation |
| title | Minimum cut model for spoken lecture segmentation |
| title_full | Minimum cut model for spoken lecture segmentation |
| title_fullStr | Minimum cut model for spoken lecture segmentation |
| title_full_unstemmed | Minimum cut model for spoken lecture segmentation |
| title_short | Minimum cut model for spoken lecture segmentation |
| title_sort | minimum cut model for spoken lecture segmentation |
| topic | Electrical Engineering and Computer Science. |
| url | http://hdl.handle.net/1721.1/51651 |
| work_keys_str_mv | AT malioutovigorigormikhailovich minimumcutmodelforspokenlecturesegmentation |