VoiceStudio
50,000Download

Speech generation · Opt-in

dots.tts

A 2B zero-shot cloning voice with 24 languages at 48 kHz. Linux and macOS only.

  • rednote-hilabBy
  • 24Languages
  • 48 kHzOutput
  • ~9 GBDownload
Get VoiceStudioEngine docsrednote-hilab/dots.tts

Can do

  • Voice cloningYes. YesSpeaks in a voice copied from a short reference clip.
  • Voice designNo. NoBuilds a new voice from a written description.
  • Emotion controlNo. NoTakes graded emotion from a clip, a vector or text.
  • Preset voicesNo. NoShips fixed voices; no reference clip needed.

Languages

24languages
Full list on the model card

Runs on

A ~9 GB checkpoint; a 12 to 16 GB CUDA GPU is the realistic target.

Hardware

EngineCUDAROCmIntel ArcApple SiliconVulkanNPUCPU
PyTorch, isolatedSupportedNot supportedNot supportedNot supportedNot supportedNot supportedSupported

Supported Needs a build Experimental

Licence

Code
Apache-2.0
Weights
Apache-2.0 (checkpoints)
  • Not available on Windows.

VoiceStudio is AGPL-3.0 and doesn't relicense model weights. Licence guide

Related

From the VoiceStudio app at 0834c8b. 27 models listed.

Quick answers