StarGAN_Voice_Test

このリポジトリはStarGANを使って多対多声質変換を目指すリポジトリです.

現状StarGANにFeatureMatching_loss,mel cepstorumを用いた学習を導入

良いハイパーパラメータが見つからない.

This repository is a repository aiming at multi-voice conversion using StarGAN.

Current FeatureMatching_loss, mel cepstrum learning introduced in StarGAN

There is no good hyper parameter found.

Name		Name	Last commit message	Last commit date
Latest commit History 22 Commits
.vscode		.vscode
assets		assets
.gitattributes		.gitattributes
.gitignore		.gitignore
IFFT_Test.py		IFFT_Test.py
README.md		README.md
StarGAN_Model.py		StarGAN_Model.py
ceps_by_pyworld.py		ceps_by_pyworld.py
cepstrum_test.py		cepstrum_test.py
converter.py		converter.py
load_datas.py		load_datas.py
main.py		main.py
model.py		model.py
module.py		module.py
phase_Test.py		phase_Test.py
test.py		test.py
trainer.py		trainer.py
util.py		util.py
wav_convert.py		wav_convert.py
wav_convert4Ceps.py		wav_convert4Ceps.py
wav_convert4phase.py		wav_convert4phase.py
wav_util.py		wav_util.py

keel-210/StarGAN_Voice_Test