Additive Margin in Contrastive Self-Supervised Frameworks to Learn Discriminative Speaker Representations.

Théo Lepage Réda Dehak

Published in: CoRR (2024)

Keyphrases

speech recognition
mid level
max margin
feature extraction
support vector
semi supervised
language model
higher level
neural network
objective function
hidden markov models
multi class
maximum margin
efficient learning
speaker verification