Cross_Utterance_Clean

This directory contains different LSTM-based cross-utterance language models (LMs)

Train/pre-train the first-level LM using vanillaLSTM.sh The model will be saved in Flvmodel directory
Train the cross-utterance LM using jointatten.sh The model will be saved in atten_<exp_no> directory

Name		Name	Last commit message	Last commit date
Latest commit History 30 Commits
data		data
error_sampling		error_sampling
AttenFlvmodel.py		AttenFlvmodel.py
ErrorSampling.py		ErrorSampling.py
L2joint_dataloader_atten.py		L2joint_dataloader_atten.py
L2model.py		L2model.py
README.md		README.md
SelfAtten.py		SelfAtten.py
dataloader.py		dataloader.py
jointatten.sh		jointatten.sh
jointforward.py		jointforward.py
jointtrain_singleseg.py		jointtrain_singleseg.py
model.py		model.py
nbestjointAtten.sh		nbestjointAtten.sh
train_with_dataloader.py		train_with_dataloader.py
vanillaLSTM.sh		vanillaLSTM.sh

BriansIDP/Cross_Utterance_Clean