mirror of https://github.com/xiph/opus.git synced 2025-05-30 23:27:42 +00:00

History

Jean-Marc Valin c4b83ae62d RDOVAE model update		2023-09-26 12:11:25 -04:00
..
packets	Remove trailing whitespace in dnn	2023-06-22 13:58:37 -07:00
rdovae	Quantizing initial state with rdovae too	2023-09-20 18:04:08 -04:00
export_rdovae_weights.py	Support for dumping LinearLayer in weight-exchange	2023-07-27 19:55:17 -04:00
fec_encoder.py	Remove trailing whitespace in dnn	2023-06-22 13:58:37 -07:00
import_rdovae_weights.py	Remove trailing whitespace in dnn	2023-06-22 13:58:37 -07:00
README.md	added pytorch implementation of RDOVAE	2022-11-23 11:02:29 +00:00
requirements.txt	Support for dumping LinearLayer in weight-exchange	2023-07-27 19:55:17 -04:00
train_rdovae.py	RDOVAE model update	2023-09-26 12:11:25 -04:00

Rate-Distortion-Optimized Variational Auto-Encoder

Setup

The python code requires python >= 3.6 and has been tested with python 3.6 and python 3.10. To install requirements run

python -m pip install -r requirements.txt

To generate training data use dump date from the main LPCNet repo

./dump_data -train 16khz_speech_input.s16 features.f32 data.s16

To train the model, simply run

python train_rdovae.py features.f32 output_folder

To train on CUDA device add --cuda-visible-devices idx.