no code implementations • 13 Feb 2024 • Théo Mariotte, Anthony Larcher, Silvio Montrésor, Jean-Hugh Thomas
A channel-number invariant loss is proposed to learn a unique feature representation regardless of the number of available microphones.
no code implementations • 7 Jun 2023 • Théo Mariotte, Anthony Larcher, Silvio Montrésor, Jean-Hugh Thomas
Pipeline systems rely on speech segmentation to extract speakers' segments and achieve robust speaker diarization.