Update README.md
Browse files
README.md
CHANGED
|
@@ -15,6 +15,7 @@ This vits model was trained on the [KTH/hungarian-single-speaker-tts](https://hu
|
|
| 15 |
The corpus consists of a single speaker, with 4515 segments extracted from [this](https://librivox.org/egri-csillagok-by-geza-gardonyi/) single LibriVox audiobook. It consists about 10 hours of audio data.
|
| 16 |
|
| 17 |
## Training
|
| 18 |
-
The model was trained on a single RTX 3090 GPU. The training took about
|
|
|
|
| 19 |
|
| 20 |
We created a Hungarian text cleaner function for this dataset, which can be found in the `text/cleaners.py` file.
|
|
|
|
| 15 |
The corpus consists of a single speaker, with 4515 segments extracted from [this](https://librivox.org/egri-csillagok-by-geza-gardonyi/) single LibriVox audiobook. It consists about 10 hours of audio data.
|
| 16 |
|
| 17 |
## Training
|
| 18 |
+
The model was trained on a single RTX 3090 GPU. The training took about 1 day for the first checkpoint (Step 93000).
|
| 19 |
+
Based on the quality of the preview model, we are aiming for 250K Steps.
|
| 20 |
|
| 21 |
We created a Hungarian text cleaner function for this dataset, which can be found in the `text/cleaners.py` file.
|