This folder contains configs and code related to the publication:
paper Dynamic Encoder Size Based on Data-Driven Layer-wise Pruning for Speech Recognition
We use RETURNN for training and our setups are based on Sisyphus.
We use models parts from i6-models
ConformerCTCModel, ConformerCTCConfig and train_step in returnn config is defined in here
ConformerCTCModel, ConformerCTCConfig and train_step in returnn config is defined in here
ConformerCTCModel, ConformerCTCConfig and train_step in returnn config is defined in here
ConformerCTCModel, ConformerCTCConfig and train_step in returnn config is defined in here
ConformerCTCModel, ConformerCTCConfig and train_step in returnn config is defined in here