AdvDCTTS (Adversarial Deep Convolutional TTS)

Prerequisite

Download the above dataset and modify the path in config.py. And then run the below command. 1st arg: signal prepro, 2nd arg: metadata (train/test split)
```
python prepro.py 1 1
```
DCTTS has two models. Firstly, you should train the model Text2Mel. I think that 20k step is enough (for only an hour). But you should train the model more and more with decaying guided attention loss.
```
python train.py text2mel <gpu_id>
```
Secondly, train the SSRN with GAN. The outputs of SSRN are many high resolution data. So training SSRN is slower than training Text2Mel
```
python gan_train.py <gpu_id>
```
After training, you can synthesize some speech from text.
```
python synthesize.py <gpu_id>
```
You can also test ssrn using the ground truth mel spectrograms.
```
python test.py <gpu_id>
```

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
__pycache__		__pycache__
assets		assets
.gitignore		.gitignore
README.md		README.md
config.py		config.py
data.py		data.py
gan_train.py		gan_train.py
layers.py		layers.py
lj_eval_idx.npy		lj_eval_idx.npy
models.py		models.py
modules.py		modules.py
network.py		network.py
prepro.py		prepro.py
synthesize.py		synthesize.py
test.py		test.py
test_sents.txt		test_sents.txt
train.py		train.py
utils.py		utils.py