Search papers, labs, and topics across Lattice.
4
0
4
Discrete audio tokens can rival traditional spectral features in speaker verification when guided by a robust knowledge distillation framework.
Uncertainty estimates in speaker recognition can be dramatically improved by incorporating both inter- and intra-speaker dynamics, leading to more reliable embeddings.
Speaker recognition accuracy improves dramatically when leveraging a U-Net-based fusion of noisy and enhanced speech, coupled with a novel training strategy.
Training on real speech prosody alone can cut speech deepfake error rates by over 70% on emotional attacks, a blindspot for current detectors.
