Abstract | ||
---|---|---|
In this paper we present the two-pass speaker diarization system that we developed for the NIST RT09s evaluation. In the first pass of our system a model for speech overlap detection is generated automatically. This model is used in two ways to reduce the diarization errors due to overlapping speech. First, it is used in a second diarization pass to remove overlapping speech from the data while training the speaker models. Second, it is used to find speech overlap for the final segmentation so that overlapping speech segments can be generated. The experiments show that our overlap detection method improves the performance of all three of our system configurations. |
Year | Venue | Keywords |
---|---|---|
2009 | INTERSPEECH 2009: 10TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2009, VOLS 1-5 | Speaker diarization, speech overlap detection, Benchmark |
Field | DocType | Citations |
Speech corpus,Speech communication,Pattern recognition,Voice activity detection,Segmentation,Computer science,Speech recognition,NIST,Speaker recognition,Speaker diarisation,Artificial intelligence | Conference | 9 |
PageRank | References | Authors |
0.61 | 6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Marijn Huijbregts | 1 | 164 | 14.92 |
David van Leeuwen | 2 | 44 | 7.06 |
Franciska de Jong | 3 | 924 | 84.89 |