DiPCo is a far-field speaker recognition benchmark derived from the publicly-available DiPCo corpus. This corpus comprises three different tasks that involve enrollment and test conditions with single- and/or multi-channel recordings. The main goal of this corpus is to foster research in far-field and multi-channel text-independent speaker recognition. Also, it can be used for other tasks such as dereverberation, denoising, and speech enhancement.
This work was granted access to the HPC resources of IDRIS under the allocation 2021-AD011013257 made by GENCI and was supported by the ANR agency (Agence Nationale de la Recherche), RoboVox project (ANR-18-CE33-0014)
The data for DiPCo are available at this address: https://dipco-sre.univ-avignon.fr/corpus.tar.gz.
The corpus is made available under the CDLA-Permissive license.