OpenVoiceV2

OpenVoiceV2

Demonstrates OpenVoice v2, a voice model from MyShell, in a Gradio app with example audio and cached outputs. It suits people who want to try the model online or read how the demo is built.

Made by myshell-ai on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: OpenVoiceV2 is a Gradio demo of OpenVoice v2, the voice model from myshell-ai. The Space names that one model. Its repository holds app.py, a checkpoint file, an examples folder, a folder of cached outputs and an MP4 file.

Who it suits: People who want to hear what OpenVoice v2 does in a browser tab before they install any voice software.

Industry | Who uses it | How they use it | Example
Audio tech | Voice developers | Try the model before building on it | One of the bundled examples
Content | Video creators | Test whether the voice output fits a project | A short narration test
Research | Speech researchers | Look at the demo's code and cached results | The cached_outputs folder
Education | Students | See a voice model run with no setup | A classroom demo

If you do not work with speech or audio, you can skip it.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Windows, macOS or Linux with Python, from a downloaded copy

Use case | Who | What they get back
Trying the model | Developers | Audio produced by OpenVoice v2
Example playback | Newcomers | The saved example results in the repository
Code study | Researchers | A short app file and a hash script to read

How to set it up:

• Python on your computer
• Python packages from requirements.txt, with Gradio 3.48.0
• No account or key is named

1. Install the Hugging Face hub package.
2. Download the Space with the hf command.
3. Move into the downloaded folder.
4. Install Gradio 3.48.0, then the requirements.
5. Start app.py.

pip install -U huggingface_hub
hf download myshell-ai/OpenVoiceV2 --repo-type space --local-dir OpenVoiceV2
cd OpenVoiceV2
pip install gradio==3.48.0
pip install -r requirements.txt
python app.py

How to use it: Online, open the Space page and use the screen there. On your own computer, start the app with the command below and use the screen it opens.

python app.py

Tool | What it does
app.py | The main Gradio file that starts the demo
examples/ | A folder of example material for the demo
cached_outputs/ | A folder of saved outputs, matched to examples by a hash script
hash_code_for_cached_output.py | A script that makes the hash codes used for cached outputs

Example: You open the Space and try one of the bundled examples. The source does not show the screen, but the demo returns audio from the voice model.

For newcomers and experts: Open the Space page and start with an example, since the repository ships its own. You will not need to install anything for that.

The Space runs on cpu-basic hardware and names the myshell-ai/OpenVoice-v2 model. The cached outputs folder and the hash script suggest that saved results are served for the examples, though the source does not say so. The Gradio version is 3.48.0, which is an older release.

You open the demo or run app.py
You pick or supply input in the Gradio screen
The OpenVoice v2 model produces audio
The result plays in the screen

Good to know:

• Hardware: cpu-basic at Hugging Face, built with Gradio 3.48.0
• License: mit, with a LICENSE file in the repository
• Cost to run: none named
• Data it reaches: the input you give it, plus the model it names
• Not for: the source names no limits

On Hugging Face, 13 November 2026

• Likes: 364
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/myshell-ai/OpenVoiceV2
• Files: https://huggingface.co/spaces/myshell-ai/OpenVoiceV2/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.