Applied ckustering for automatic speaker-based segmentation of audio material

Authors

  • O. Pietquin Faculté Polytechnique de Mons
  • P. Couvreur Université de Mons-Hainaut

Abstract

In this paper, we introduce an algorithm dedicated to speaker-based segmentation of audio material. The algorithm consists in two distinct procedures namely splitting and merging. Its performance is assessed on broadcast news recordings provided by the
British Broadcast Corporation (BBC). Results show that the splitting is performed with high accuracy and low missed detection rate while the merging procedure provides satisfying results.

Downloads

Published

2001-06-01

How to Cite

Pietquin, O., & Couvreur, P. (2001). Applied ckustering for automatic speaker-based segmentation of audio material. JORBEL - Belgian Journal of Operations Research, Statistics, and Computer Science, 40(3-4), 69–81. Retrieved from https://www.orbel.be/jorbel/index.php/jorbel/article/view/346

Issue

Vol. 40 No. 3-4 (2000)

Section

Articles