End-to-end version of lattice-free MMI (LF-MMI or chain model) implemented in PyTorch.
TODO: regular version of LF-MMI.

• August 2020: GPU computation for graphs in log domain (recommended for numerator graphs)
• April 2020: Support unequal length sequences within a minibatch
• April 2020: Examples of using PyChain: Espresso and pychain-example
• January 2020: GPU computation for both denominator and numerator graphs

• PyTorch version >= 1.4.0

pip install kaldi_io
git clone https://github.com/YiwenShaoStephen/pychain.git
cd pychain
make

Whenever you update or modify any none-python codes (e.g. .c or .cu) in pychain, you need to re-compile it by

make pychain

"PyChain: A Fully Parallelized PyTorch Implementation of LF-MMI for End-to-End ASR", Yiwen Shao, Yiming Wang, Daniel Povey and Sanjeev Khudanpur (pdf)