eastonyi Goto Github PK

asr-tf1

ASR recipes using TF1.x

audio2text

bert-as-language-model

bert as language model, fork from https://github.com/google-research/bert

cmmlu

cyclegan

cycleGAN for image transformation (original paper) and text transformation (CipherGAN)

eastoncode

the models and utils designed for general use

end-to-end_asr_pytorch

Implements of CTC, Speech-Transformer and CIF for end-to-end speech recognition with pytorch

espnet

End-to-End Speech Processing Toolkit

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

libri-light

dataset for lightly supervised training using the librivox audio book recordings. https://librivox.org/.

lm_eval

openasr

A pytorch based end2end speech recognition system.

openseq2seq

Toolkit for efficient experimentation with various sequence-to-sequence models

reinforcementlearning

RL demos. Mainly implement based on pure Python. Neural Network (mainly Tensorflow) is only used as value estimator. Long-Term Support

tf-mnist

transformer

A TensorFlow Implementation of the Transformer: Attention Is All You Need

unsupervised-asr

unsupervised ASR (mainly phone classifier) using EODM and GAN

wav2vec

a simplified version of wav2vec(1.0, vq, 2.0) in fairseq

wfst

used for speech recognition

write-rnn-tensorflow

Generative Handwriting using LSTM Mixture Density Network with TensorFlow