Text-to-Speech
WaveGen
Safetensors
PyTorch
English
Chinese
WaveGen
TTS
zero-shot
voice-cloning
audio
multilingual
Instructions to use LeeSangHoon/WaveGen with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- WaveGen
How to use LeeSangHoon/WaveGen with WaveGen:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Upload README.md
Browse files
README.md
CHANGED
|
@@ -53,8 +53,8 @@ weights and BF16 CUDA inference. The original test prompts are used.
|
|
| 53 |
|
| 54 |
| Model | CER (%) ↓ | SIM ↑ | UTMOS ↑ |
|
| 55 |
|---|---:|---:|---:|
|
| 56 |
-
| WaveGen (Pre-trained, 1300k EMA) | 1.
|
| 57 |
-
| WaveGen V-Spec (1400k EMA) | 1.
|
| 58 |
|
| 59 |
WER is reported for English and CER for Chinese. Error rates are percentages;
|
| 60 |
all metrics are macro-averaged over utterances. Values are shown without
|
|
|
|
| 53 |
|
| 54 |
| Model | CER (%) ↓ | SIM ↑ | UTMOS ↑ |
|
| 55 |
|---|---:|---:|---:|
|
| 56 |
+
| WaveGen (Pre-trained, 1300k EMA) | 1.75 | 0.805 | 2.87 |
|
| 57 |
+
| WaveGen V-Spec (1400k EMA) | 1.65 | 0.783 | 2.79 |
|
| 58 |
|
| 59 |
WER is reported for English and CER for Chinese. Error rates are percentages;
|
| 60 |
all metrics are macro-averaged over utterances. Values are shown without
|