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

Théo Lepage Réda Dehak

Published in: Odyssey (2024)

Keyphrases

training set
speech recognition
higher level
mid level
objective function
semi supervised
unsupervised learning
max margin
feature selection
feature extraction
bayesian network classifiers
speaker verification
speaker recognition