Awesome Open

Project record

OpenVoice

OpenVoiceInstant voice cloning and audio foundation model.MIT
Field notes / 525OpenVoice
I / Function

Instant voice cloning and audio foundation model.

OpenVoice generates speech using the tone color of a reference voice, with controls for style characteristics such as emotion, accent, rhythm and intonation. Its second version natively supports English, Spanish, French, Chinese, Japanese and Korean. Researchers can run the Python and PyTorch implementation on Linux using supplied checkpoints, while hosted demonstrations provide a separate way to try voice cloning.

OpenVoice README / OpenVoice usage guide · Reviewed 2026-10-27

A folded acoustic reed carries one patterned membrane through two differently shaped sound openings.
Awesome Open* · Field recordOpenVoiceOP-4C898016 · Folio 525Domain: AI & Language Models
  1. 01 / Feature noteCross-language voice cloning

    Use a reference voice to generate speech in another supported language.

    Feature source ↗
  2. 02 / Feature noteStyle controls

    Adjust characteristics including emotion, accent, rhythm, pauses and intonation.

    Feature source ↗
II / Repository record

GitHub stars

38,136

Detected license

MIT

Repository created

2023-11-29

Repository push

2025-04-19

III / Observations & evidence

GitHub stars

Unknown

i
Window mismatch; awaiting refresh

Cached history uses a different month window; chart withheld until aligned.

Latest attempt: ok · Checked 2026-10-15T06:49:08.002085+00:00 · Last successful fetch 2026-10-15T06:49:08.002085+00:00 · Provider source

Hacker News stories

Unknown

i
Not recorded

OpenSSF Scorecard

Unknown

i
Unavailable

No result returned.

Checked 2026-10-26T23:54:49Z · Latest attempt: not found · Provider source

IV / Release record

GitHub release

Unknown

i
Unavailable

No result returned.

Checked 2026-11-04T04:04:05Z · Latest attempt: not found · Provider source

V / Where it runs
Library / SDKLocal browser interface
Build from source
Linux
i

Run Install Documented platforms

Reviewed 2026-10-31 · Kanzler (AI) · Approved evidence record

Run modes
  • Library / SDK

    pip install -e . installs the cloned package. Demos are Jupyter notebooks.

    OpenVoice usage

  • Local browser interface

    A local Gradio demo is launched with python -m openvoice_app --share.

    OpenVoice usage

Install methods
  • Build from source

    On Linux: conda with Python 3.9, git clone, and pip install -e . Checkpoint zips are downloaded separately.

    OpenVoice usage

Documented platforms
  • Linux

    The project labels the Linux procedure as the install for developers. Windows and Docker guides on that page are marked unofficial community guides.

    OpenVoice usage

Resource and setup notes
  • Minimum: Python 3.9.

    conda create -n openvoice python=3.9. The page also says the reader should know Linux, Python, and PyTorch.

    OpenVoice usage

  • Constraint: OpenVoice V2 also installs MeloTTS and downloads unidic.

    pip install git+https://github.com/myshell-ai/MeloTTS.git and python -m unidic download.

    OpenVoice usage