OpenVoice myshell-ai/OpenVoice

MyShell's instant voice cloning model with tone-colour transfer, style control and cross-lingual cloning; MIT code and weights.

Creator kit + Compare GitHub ↗ Install guide (official) ↗

OpenVoice at a glance

Checked 15 Oct 2026

OpenVoice clones a reference voice's tone colour and speaks it in other languages; V2 (April 2024) natively supports English, Spanish, French, Chinese, Japanese and Korean and uses MeloTTS (MIT) as the base speaker. Code and V1/V2 weights are MIT and the README states free commercial use. Official install is Linux only (conda, manual checkpoint download); Windows and Docker guides are community-contributed. No hardware requirements are documented. Last push April 2025.

Cite this pageRepoQuarry. “OpenVoice.” Checked 15 Oct 2026. https://repoquarry.dev/projects/openvoice

How much of ElevenLabs does it replace?

2 of 4 jobs. Instant zero-shot cloning in six native languages via MeloTTS; official setup is Linux and notebook-based.

Speech generation Partial replacement

Voice cloning Partial replacement

Speech recognition Not covered

Voice agents Not covered

For speech recognition, voice agents, see other ElevenLabs alternatives. You may also need to build an interface, an API server and hosting yourself.

How much of Descript does it replace?

2 of 3 jobs. Voice cloning and text-to-speech only; no transcript-based audio or video editing.

Transcript-based editing Not covered

Voice cloning Partial replacement

Speech generation Partial replacement

For transcript-based editing, see other Descript alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction

QuestionAnswerSourceChecked

Runs on my machine? Yes, Python package with notebooks and a local Gradio demo Launch a local gradio demo with python -m openvoice_app --share. github.com · USAGE.md ↗ 1d ago

Code licence MIT OpenVoice V1 and V2 are MIT Licensed. Free for both commercial and research use. github.com · LICENSE ↗ 1d ago

Model weights licence MIT (V1 and V2 checkpoints; Hugging Face model card myshell-ai/OpenVoiceV2) Starting from April 2024, both V2 and V1 are released under MIT License. Free for commercial use. huggingface.co · OpenVoiceV2 ↗ 1d ago

Commercial use Yes, MIT code and weights; V2 also needs MeloTTS, which is MIT OpenVoice V1 and V2 are MIT Licensed. Free for both commercial and research use. github.com · OpenVoice#license ↗ 1d ago

Paid services needed? None required Download the checkpoint from here and extract it to the checkpoints_v2 folder. github.com · USAGE.md ↗ 1d ago

Hardware Not documented Know the answer? —

Operating systems Official install is Linux only; Windows and Docker guides are unofficial community contributions; macOS not mentioned This section is only for developers and researchers who are familiar with Linux, Python and PyTorch. github.com · USAGE.md ↗ 1d ago

Setup effort Medium, conda env, pip install -e ., manual checkpoint zip download, plus MeloTTS and unidic for V2 pip install git+https://github.com/myshell-ai/MeloTTS.git github.com · USAGE.md ↗ 1d ago