SVS_system
A system works on singing voice synthesis
Environment Install
cd tools
make KALDI=your_kaldi_dir
and then install torch by conda
For usage:
. tools/env.sh
Followings are my sample environment:
python version: 3.7.3 (default, Mar 27 2019, 22:11:17) [GCC 7.3.0]
pytorch version: 1.1.0
cupy version: 7.3.0
cuda version: 9000
cudnn version: 7501
Running Instruction
For example:
cd egs/public_dataset/kiritan
sh train.sh
- For CLSP User, using clsp_wrapper to use qsub.
- For other user, using train.sh or infer.sh to run.
Please refer to configuration file (e.g. train.yaml) for parameter.