k2-fsa (Next-gen Kaldi) · CC BY-NC 4.0
OmniVoice
A 0.6B voice model covering 600+ languages with cloning. The weights are non-commercial.
How much memory it needs
Memory to run each model, from its model card or the size of its files. Smaller versions trade some quality for less memory.
| Model | Version | Needs | Business use |
|---|---|---|---|
| OmniVoice | FP32 | 3.3 GB | Not allowed |
Which computers run it
The best model in this group for each machine.
| Computer | Best version | Fit |
|---|---|---|
| Laptop or PC, no graphics card, 32 GB | OmniVoice FP32 | Runs great |
| RTX 4060 · 8 GB | OmniVoice FP32 | Runs great |
| RTX 3060 · 12 GB | OmniVoice FP32 | Runs great |
| RTX 5060 Ti · 16 GB | OmniVoice FP32 | Runs great |
| RTX 4090 · 24 GB | OmniVoice FP32 | Runs great |
| RTX 5090 · 32 GB | OmniVoice FP32 | Runs great |
| Mac with M4 · 16 GB | OmniVoice FP32 | Runs great |
| Mac with M4 Pro · 48 GB | OmniVoice FP32 | Runs great |
| Mac with M5 Max · 128 GB | OmniVoice FP32 | Runs great |
| Ryzen AI Max+ 395 · 128 GB | OmniVoice FP32 | Runs great |
| DGX Spark · 128 GB | OmniVoice FP32 | Runs great |
| RTX PRO 6000 · 96 GB | OmniVoice FP32 | Runs great |
| Mac Studio with M5 Ultra · 256 GB | OmniVoice FP32 | Runs great |
| Mac Studio with M3 Ultra · 512 GB | OmniVoice FP32 | Runs great |
What it is good and bad at
Good at
- Huge language coverage
Not good at
- Non-commercial (CC BY-NC)
Start it
- OmniVoice: runs in omnivoice (PyTorch) (also GGUF community ports). Model card