AI RESEARCH
DELULU: Discriminative Embedding Learning Using Latent Units for Speaker-Aware Self-Trained Speech Foundational Model
arXiv CS.CL
•
ArXi:2510.17662v2 Announce Type: replace-cross Self-supervised speech models have achieved remarkable success on content-driven tasks, yet they remain limited in capturing speaker-discriminative features critical for verification, diarization, and profiling applications. We