logo Idiap Research Institute        
 [BibTeX] [Marc21]
Short-Term Spatio-Temporal Clustering of Sporadic and Concurrent Events
Type of publication: Idiap-RR
Citation: lathoud-rr-04-14
Number: Idiap-RR-14-2004
Year: 2004
Institution: IDIAP
Address: Martigny, Switzerland
Note: Published in ``Proceedings of the 2004 ICASSP-NIST Meeting Recognition Workshop''
Abstract: Accurate detection and segmentation of spontaneous multi-party speech is crucial for a variety of applications, including speech acquisition and recognition, as well as higher-level event recognition. However, the highly sporadic nature of spontaneous speech makes this task difficult. Moreover, multi-party speech contains many overlaps. We propose to attack this problem as a multitarget tracking task, using location cues only. In order to best deal with high sporadicity, we propose a novel, generic, short-term clustering algorithm that can track multiple objects for a low computational cost. The proposed approach is online, fully deterministic and can run in real-time. In an application to real meeting data, the algorithm produces high precision speech segmentation.
Userfields: ipdinar={2004}, ipdmembership={speech, vision}, language={English},
Keywords:
Projects Idiap
Authors Lathoud, Guillaume
McCowan, Iain A.
Odobez, Jean-Marc
Crossref by lathoud04a
Added by: [UNK]
Total mark: 0
Attachments
  • rr-04-14.pdf
  • rr-04-14.ps.gz
Notes