Description
new month. new voice model. this voice model may sound a bit worse than the old dizzy model i made earlier trained on 16 minutes of audio data from her recent streams. (unwa's mel roformer + uvr de-reverb + rx de-click + de-plosive + eq) pitch extraction: rmvpe steps: 18.9k / 19k steps batch size: 7 pretrain: klm 4.1 / 32k huggingface:
Comments
No comments yet. Start the conversation!
Add a comment
Samples
New
Classic
1. Singing
Male
English
2. Singing
Female
English
3. Singing (Dry)
Female
English
4. Singing (High)
Female
English
5. Singing 2
Male
English
6. Singing (Dry)
Male
English
7. Singing (Dry, High)
Male
English
Pitch