Update README.md
Browse files
README.md
CHANGED
|
@@ -15,3 +15,5 @@ This repository is experimental, expect some bugs and some hardcoded params.
|
|
| 15 |
The default setting is 44.1khz - 128 Mel bin. if you want to change it to 24khz, copy the config from HiFTnet (make sure to copy its pitch extractor, both the model + the checkpoint.), then change 128 to 80 in LN-384 of the models.py. then uncomment the "multiscale_subband_cfg" for the 24khz version.
|
| 16 |
|
| 17 |
Huge Thanks to [Johnathan Duering](https://github.com/duerig) for his help. I mostly implemented this based on his [STTS2 Fork](https://github.com/duerig/StyleTTS2/tree/main).
|
|
|
|
|
|
|
|
|
| 15 |
The default setting is 44.1khz - 128 Mel bin. if you want to change it to 24khz, copy the config from HiFTnet (make sure to copy its pitch extractor, both the model + the checkpoint.), then change 128 to 80 in LN-384 of the models.py. then uncomment the "multiscale_subband_cfg" for the 24khz version.
|
| 16 |
|
| 17 |
Huge Thanks to [Johnathan Duering](https://github.com/duerig) for his help. I mostly implemented this based on his [STTS2 Fork](https://github.com/duerig/StyleTTS2/tree/main).
|
| 18 |
+
|
| 19 |
+
**This is highly experimental, I have not conducted a full session training. I just tested that the loss goes down and the eval samples sound reasonable for ~10K steps of minimal training.**
|