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. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/ICASSP.2007.366652 | ICASSP (1) |
Keywords | Field | DocType |
tri-model bayesian information criterion,trimodel bayesin information criterion,bayes methods,audio segmentation,segment detection,data balance ratio,audio signal processing,acoustic change detection,unsupervised audio segmentation approach,bayesian methods,indexing,speech recognition,feature extraction,loudspeakers,bayesian information criterion,change detection,tv,motion pictures | Computer vision,Bayesian information criterion,Scale-space segmentation,Pattern recognition,Computer science,Segmentation,Search engine indexing,Segmentation-based object categorization,Feature extraction,Artificial intelligence,Audio signal processing,Bayesian probability | Conference |
Volume | ISSN | ISBN |
1 | 1520-6149 | 1-4244-0727-3 |
Citations | PageRank | References |
1 | 0.35 | 5 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yunfeng Du | 1 | 1 | 0.35 |
Wei Hu | 2 | 182 | 14.17 |
Yonghong Yan | 3 | 656 | 114.13 |
Tao Wang | 4 | 238 | 23.70 |
Yimin Zhang | 5 | 359 | 28.66 |