82,040 undefined

· undefined

FunASR SenseVoice

undefined

Alibaba DAMOundefined
50+undefined

modelscope/FunASR

50+

50+

PyTorch
PyTorchCUDACPU
Sentence-level timestamps only.

	CUDA	ROCm	Intel Arc	Apple Silicon	Vulkan	NPU	CPU
PyTorch

: : :

iic/SenseVoiceSmall ·

: NOASSERTION
: iic/SenseVoiceSmall






: 2026-10-03

Runtime uses the FunASR/ModelScope identifier. The matching Hugging Face API returned 401; no governing weight terms have been verified.

undefined

: 626f87eb82dfe114fe00f08ebc4754cb6522d23368d5206654a0ccdb6dd8f2d2
: a47e3e0c421a39e1c16a0dfb79fd4224bbfe5d2ea8b254c02b7fa2e4b53fe7a3
: 06c6e077f0fc35149efefc3561e9be5ae835d916
: 090cd5b58cf771def46f64352a0de72c2432fc56

undefined. undefined

Is it really free?
The desktop app is open source under AGPL-3.0, so yes. Pro and Lifetime are a separate, closed-source build with a paid Commercial License and extra features, for teams that need commercial terms.
Licence guide
Does it phone home?
No account, no cloud. It only goes online when you ask it to: checking for updates or downloading a model.
Privacy
Do I need a fancy GPU?
Nope. About 8 GB of RAM is enough to start. A GPU just makes everything faster.
System needs
Can I clone any voice?
Only with the owner's say-so. Impersonation and fraud are off the table.
The rules
Changed my mind. Refund?
undefined
Refund policy
Where do I get help?
Write to us or open an issue. Real inboxes, answered within two business days.
Contact