A modern replacement for Redis and Memcached

Tiny configuration for Triton Inference Server

[NeurIPS'22] Squeezeformer: An Efficient Transformer for Automatic Speech Recognition

LightSeq: A High Performance Library for Sequence Processing and Generation

Extensions to YAML syntax for better python interaction

A PyTorch-based Speech Toolkit

Training RNNs as Fast as CNNs (https://arxiv.org/abs/1709.02755)

PORORO: Platform Of neuRal mOdels for natuRal language prOcessing