Audio Segmentation via Tri-Model Bayesian Information Criterion

Yunfeng Du, Wei Hu, Yonghong Yan, Tao Wang, Yimin ZhangICASSP 2007 · OpenReview

Abstract

This paper addresses the problem of audio segmentation in practical media (e.g. TV series, movies and etc.) which usually consists of segments in various lengths with quite a portion of short ones. An unsupervised audio segmentation approach is presented, including a segmentation-stage to detect potential acoustic changes, and a refinement-stage to refine these candidate changes by a tri-model Bayesian information criterion. Experiments show that the proposed approach has good detectability of short segments and the novel tri-model BIC effectively improves the overall segmentation performance.
BibTeX
@inproceedings{du2007audio,
  author = {Yunfeng Du and Wei Hu and Yonghong Yan and Tao Wang and Yimin Zhang},
  title = {{Audio Segmentation via Tri-Model Bayesian Information Criterion}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2007},
  url = {https://openreview.net/forum?id=KllM2IjkuJ},
}

Loading related content...

Figure