1*a58d3d2aSXin Li# Rate-Distortion-Optimized Variational Auto-Encoder 2*a58d3d2aSXin Li 3*a58d3d2aSXin Li## Setup 4*a58d3d2aSXin LiThe python code requires python >= 3.6 and has been tested with python 3.6 and python 3.10. To install requirements run 5*a58d3d2aSXin Li``` 6*a58d3d2aSXin Lipython -m pip install -r requirements.txt 7*a58d3d2aSXin Li``` 8*a58d3d2aSXin Li 9*a58d3d2aSXin Li## Training 10*a58d3d2aSXin LiTo generate training data use dump date from the main LPCNet repo 11*a58d3d2aSXin Li``` 12*a58d3d2aSXin Li./dump_data -train 16khz_speech_input.s16 features.f32 data.s16 13*a58d3d2aSXin Li``` 14*a58d3d2aSXin Li 15*a58d3d2aSXin LiTo train the model, simply run 16*a58d3d2aSXin Li``` 17*a58d3d2aSXin Lipython train_rdovae.py features.f32 output_folder 18*a58d3d2aSXin Li``` 19*a58d3d2aSXin Li 20*a58d3d2aSXin LiTo train on CUDA device add `--cuda-visible-devices idx`. 21*a58d3d2aSXin Li 22*a58d3d2aSXin Li 23*a58d3d2aSXin Li## ToDo 24*a58d3d2aSXin Li- Upload checkpoints and add URLs 25