search

hirofumi0810 / neural_sp

End-to-end ASR/LM implementation with PyTorch
Apache License 2.0
589 stars 140 forks source link

Fix LM inference #309

Closed hirofumi0810 closed 3 years ago

hirofumi0810 commented 3 years ago

Fix batch_size when calculating PPL.