LongCat Audio Tokenizer and Detokenizer

A multi-voice TTS system trained with an emphasis on quality

Hackable and optimized Transformers building blocks, supporting a composable construction.

CUDA-Warp RNN-Transducer

Automatic differentiation with weighted finite-state transducers.

Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

PyTorch bindinga for Baidu's Warp-CTC

Connectionist Temporal Classification (CTC) Automatic Speech Recognition

Custom decoders for Kaldi

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

Some meaningless nscripter tools.

Some meaningless nscripter tools.

📙 中华新华字典数据库。包括歇后语，成语，词语，汉字。

Lingvo

Tacotron 2 - PyTorch implementation with faster-than-realtime inference

tensorboard for pytorch (and chainer, mxnet, numpy, ...)

结巴中文分词

"结巴"中文分词的C++版本

A toolkit for making real world machine learning and data analysis applications in C++

PyTorch implementations of Generative Adversarial Networks.

A CV toolkit for my papers.

Pytorch Bindings for warp-ctc

A Python wrapper for Kaldi

Keras implementations of Generative Adversarial Networks.

Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding

Bitcoin Core integration/staging tree

CUDA Templates and Python DSLs for High-Performance Linear Algebra

The official repository of the Eesen project

🐍 Python library for n-gram models in ARPA format

Rasa Core is now part of the Rasa repo: An open source machine learning framework to automate text-and voice-based conversations