Description
Stats for Nerds Experiment Name: Merchant Model Architecture: v2 Target Sample Rate: 48k Speaker ID: 0 Pitch Extraction Algorithm: RMVPE Crepe Hop Length: 64 Pitch Guidance: Yes Pretrained Model: original Dataset Origin: Extracted from game files Dataset Length (hh/mm/ss): 00:15:07 Dataset File Type: .wav Dataset Sample Rate: 48k Dataset Channels: Mono Dataset Bit Rate: 768.0 kbps Dataset Bit Depth: 32 bit Batch Size: 8 Steps: 5200 Epochs: 100 Overtraining Detected: Yes Tensor Board Settings: Audacity Settings: Audacity was not used.
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
More models by