End-to-end-ASR-Pytorch
End-to-end-ASR-Pytorch copied to clipboard
Alexander-H-Liu
This is an open source project (formerly named Listen, Attend and Spell - PyTorch Implementation) for end-to-end ASR implemented with Pytorch, the well known deep learning toolkit.
Ocd
implement Optimal Completion Distillation. add a new config named libri_ocd_example.yaml which enables ocd training. Not well tested. Might have bugs inside. temperature annealing not yet implemented. currently equals to 1e-8...
In Step 0, you mentioned we can use one of the following options phoneme/char/subword/word. But when I choose "word" instead of "subword". The encoding doesn't recognize it. The error is...
Hello, I want to ask if the pure ctc training and decoding is tested and correct? Thnaks.
Metadata
Owner
Metadata
This is an open source project (formerly named Listen, Attend and Spell - PyTorch Implementation) for end-to-end ASR implemented with Pytorch, the well known deep learning toolkit.