|
--- |
|
license: mit |
|
language: |
|
- ja |
|
- vi |
|
metrics: |
|
- accuracy |
|
pipeline_tag: audio-to-audio |
|
tags: |
|
- rvc |
|
- music |
|
- art |
|
--- |
|
# <center> RVC Genshin Impact Japanese Voice Model<br /> |
|
![model-cover.png](https://huggingface.co/ArkanDash/rvc-genshin-impact/resolve/main/model-cover.png) |
|
|
|
# About Retrieval based Voice Conversion (RVC) |
|
Learn more about Retrieval based Voice Conversion in this link below:<br /> |
|
[RVC WebUI](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI) |
|
|
|
# How to use? |
|
Download the prezipped model and put to your RVC Project. |
|
|
|
Model test: [Google Colab](https://colab.research.google.com/drive/110kiMZTdP6Ri1lY9-NbQf17GVPPhHyeT?usp=sharing) / [RVC Models New](https://huggingface.co/spaces/ArkanDash/rvc-models-new) (Which is basically the same but hosted on spaces) |
|
|
|
|
|
## <center> INFO <br /> |
|
Model Created by ArkanDash <br /> |
|
The voice that was used in this model belongs to Hoyoverse. |
|
|
|
The voice I make to make this model was ripped from the game (3.7). <br /> |
|
Total Models: 29 Models (19 V1 Models & 10 V2 Models) |
|
|
|
### V1 Model <br /> |
|
This was trained on Original RVC.<br /> |
|
Pitch Extract using Harvest.<br /> |
|
This model was trained with 100 epochs, 10 batch sizes, and a 40K sample rate (some models had a 48k sample rate).<br /> |
|
Every V1 model was trained more or less around 30 minutes of character voice. |
|
|
|
I may exclude some models to higher epochs due to the low duration of the character's voice.<br /> |
|
- Klee 150 Epochs |
|
- Fischl 150 Epochs |
|
|
|
### (New) V2 Model <br /> |
|
This was trained on Mangio-Fork RVC.<br /> |
|
Pitch Extract using Crepe.<br /> |
|
This model was trained with 100 epochs, 8 batch sizes, and a 48K sample rate. (some models had a 40k sample rate).<br /> |
|
Every V2 model was trained more or less around 60 minutes of character voice. |
|
|
|
Other request:<br /> |
|
- Greater Lord Rukkhadevata: 750 Epochs, 16 Batch size, 48k Sample rate. (10 minutes dataset) |
|
- Charlotte: 400 Epochs, 16 Batch size, 48k Sample rate. (18 minutes dataset) |
|
|
|
Note: |
|
- For faruzan, somehow the index file is smaller, But it output a log when training here: <br /> |
|
`Converged (lack of improvement in inertia) at step 1152/48215` <br /> |
|
Let me know if you have an issue. |