Semi-Supervised Training with Pseudo-Labeling for End-To-End Neural Diarization.
Yuki TakashimaYusuke FujitaShota HoriguchiShinji WatanabeLeibny Paola García-PereraKenji NagamatsuPublished in: Interspeech (2021)
Keyphrases
- end to end
- semi supervised
- supervised learning
- active learning
- ad hoc networks
- wireless ad hoc networks
- label propagation
- admission control
- neural network
- congestion control
- network architecture
- multipath
- application layer
- labeled data
- semi supervised learning
- training set
- pairwise
- unlabeled data
- scalable video
- high bandwidth
- partially labeled