Unsupervised methods for speaker diarization
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011.
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | eng |
| Published: |
Massachusetts Institute of Technology
2011
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/66478 |
| _version_ | 1811088927505776640 |
|---|---|
| author | Shum, Stephen (Stephen Hin-Chung) |
| author2 | James R. Glass and Najim Dehak. |
| author_facet | James R. Glass and Najim Dehak. Shum, Stephen (Stephen Hin-Chung) |
| author_sort | Shum, Stephen (Stephen Hin-Chung) |
| collection | MIT |
| description | Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011. |
| first_indexed | 2024-09-23T14:10:08Z |
| format | Thesis |
| id | mit-1721.1/66478 |
| institution | Massachusetts Institute of Technology |
| language | eng |
| last_indexed | 2024-09-23T14:10:08Z |
| publishDate | 2011 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/664782019-04-10T18:00:42Z Unsupervised methods for speaker diarization Shum, Stephen (Stephen Hin-Chung) James R. Glass and Najim Dehak. 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, 2011. Cataloged from PDF version of thesis. Includes bibliographical references (p. 93-95). Given a stream of unlabeled audio data, speaker diarization is the process of determining "who spoke when." We propose a novel approach to solving this problem by taking advantage of the effectiveness of factor analysis as a front-end for extracting speaker-specific features and exploiting the inherent variabilities in the data through the use of unsupervised methods. Upon initial evaluation, our system achieves state-of-the art results of 0.9% Diarization Error Rate in the diarization of two-speaker telephone conversations. The approach is then generalized to the problem of K-speaker diarization, for which we take measures to address issues of data sparsity and experiment with the use of the von Mises-Fisher distribution for clustering on a unit hypersphere. Our extended system performs competitively on the diarization of conversations involving two or more speakers. Finally, we present promising initial results obtained from applying variational inference on our front-end speaker representation to estimate the unknown number of speakers in a given utterance. by Stephen Shum. S.M. 2011-10-17T21:31:07Z 2011-10-17T21:31:07Z 2011 2011 Thesis http://hdl.handle.net/1721.1/66478 756462731 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 95 p. application/pdf Massachusetts Institute of Technology |
| spellingShingle | Electrical Engineering and Computer Science. Shum, Stephen (Stephen Hin-Chung) Unsupervised methods for speaker diarization |
| title | Unsupervised methods for speaker diarization |
| title_full | Unsupervised methods for speaker diarization |
| title_fullStr | Unsupervised methods for speaker diarization |
| title_full_unstemmed | Unsupervised methods for speaker diarization |
| title_short | Unsupervised methods for speaker diarization |
| title_sort | unsupervised methods for speaker diarization |
| topic | Electrical Engineering and Computer Science. |
| url | http://hdl.handle.net/1721.1/66478 |
| work_keys_str_mv | AT shumstephenstephenhinchung unsupervisedmethodsforspeakerdiarization |