Focal speakers: a speaker selection method able to deal with heterogeneous similarity criteria

Sacha Krstulovic, Frédéric Bimbot, Delphine Charlet, Olivier Boëffard

In the context of the Neologos speech database creation project, we have studied several methods for the selection of representative speaker recordings. These methods operate a selection by optimizing a quality criterion defined in various speaker similarity modeling frameworks. The obtained selections can be cross-validated in the modeling frameworks which were not used for the optimization. The compared methods include K-Medians clustering, Hierarchical clustering, and a new method called the selection of Focal Speakers. Among these, only the new method is able to solve the joint optimization, across all the modeling frameworks, of the selection of representative speakers.

doi: 10.21437/Interspeech.2005-655

Cite as: Krstulovic, S., Bimbot, F., Charlet, D., Boëffard, O. (2005) Focal speakers: a speaker selection method able to deal with heterogeneous similarity criteria. Proc. Interspeech 2005, 3057-3060, doi: 10.21437/Interspeech.2005-655

@inproceedings{krstulovic05_interspeech,
  author={Sacha Krstulovic and Frédéric Bimbot and Delphine Charlet and Olivier Boëffard},
  title={{Focal speakers: a speaker selection method able to deal with heterogeneous similarity criteria}},
  year=2005,
  booktitle={Proc. Interspeech 2005},
  pages={3057--3060},
  doi={10.21437/Interspeech.2005-655}
}