Lean and sweet spot
Pick a pitch, compare the Quality Score, then Run. Hold Ctrl or Cmd to select multiple pitches.
Updated model as of 5\28\24. Improvement from the last model and the dataset are now merged with talking and singing.
Tip: De-Reverb your audio first before inferencing.
Dataset Info:
- Songs and Dialogue from Frozen 2 and Olaf's Frozen Adventure
- 40Khz Sample Rate
- 6 Minutes Length
Training Settings:
- Mangio Crepe
- 8 Batch Size
- 1400 Epochs
- 40K Sample Rate
- Normal V2 Pretrain
Model Link:
https://huggingface.co/SCRFilms/AnnaFrozenV4/resolve/main/anna_v6_mangio_overtrain_1400.zip?download=true
Tags: No tags available
Download Link: https://huggingface.co/SCRFilms/AnnaFrozenV4/resolve/main/anna_v6_mangio_overtrain_1400.zip?download=true Hosted on Hugging Face
Convert this model in EasyAIVoice
Pick a pitch, compare the Quality Score, then Run. Hold Ctrl or Cmd to select multiple pitches.
Selected-sample labels, grouped by what they describe.
Nearby voices from generated samples. Pick a pitch in the Sample Deck to refresh the matches.