Update README.md
Browse files
README.md
CHANGED
|
@@ -7,6 +7,9 @@
|
|
| 7 |
* StyleSpeech is based on 16k Hz sampling rate, and there is no available 16k Hz multi-speaker vocoder.
|
| 8 |
* Thus I train this vocoder from scratch using Libri-TTS train-100 hour dataset. The training pipeline is the same as the official MelGAN (https://github.com/descriptinc/melgan-neurips).
|
| 9 |
* The synthesized sounds are close to the official demo with good quality.
|
|
|
|
|
|
|
|
|
|
| 10 |
#### Training Details
|
| 11 |
* GPU: RTX 2080Ti
|
| 12 |
* Training epoch: 3000
|
|
|
|
| 7 |
* StyleSpeech is based on 16k Hz sampling rate, and there is no available 16k Hz multi-speaker vocoder.
|
| 8 |
* Thus I train this vocoder from scratch using Libri-TTS train-100 hour dataset. The training pipeline is the same as the official MelGAN (https://github.com/descriptinc/melgan-neurips).
|
| 9 |
* The synthesized sounds are close to the official demo with good quality.
|
| 10 |
+
#### Usage
|
| 11 |
+
* Please follow the official MelGAN (https://github.com/descriptinc/melgan-neurips) to load pre-trained checkpoint and convert your mel-spectrogram back to the waveform.
|
| 12 |
+
|
| 13 |
#### Training Details
|
| 14 |
* GPU: RTX 2080Ti
|
| 15 |
* Training epoch: 3000
|