Applio by IAHispano

Applio by IAHispano

A voice conversion tool with a browser interface, built for ease of use and quality, with plugin support. For artists, developers and researchers who convert only audio they have the right to use.

Made by IAHispano on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: Applio is a voice conversion tool from IAHispano that changes the voice in audio, with a Gradio interface that opens in your browser. It aims at simplicity, quality and performance, and plugins and configuration options let you adapt it to your project. It builds on VoiceConversionWebUI.

Who it suits: Artists, developers and researchers who want to convert voices through a simple interface and who own the audio or have permission to use it.

Industry | Who uses it | How they use it | Example
Music | Artists | Transform voices in audio they own or may use | Convert a voice recording they made themselves
Software | Developers | Extend the tool with plugins and configurations | Add a plugin that a project needs
Research | Audio researchers | Train voice models and follow the training in TensorBoard | Watch the charts of a training run

Anyone who cannot get permission for the voices they process should skip it, since the terms of use require ownership or explicit permission.

Where to use it: It runs on your own computer, and the interface opens in a web browser.

• Windows, through the .bat scripts
• Linux and macOS, through the .sh scripts
• Your default web browser, where the Gradio interface opens
• Google Colab notebooks, with or without the interface, which the card links to

Use case | Who | What they get back
Voice conversion | Artists | A converted audio file in a different voice
Model training | Researchers | Training runs you can follow in TensorBoard
Plugin work | Developers | Extra features added through the plugin system

How to set it up: You run one install script for your system, then start the app.

• A Windows, Linux or macOS computer
• The project files from the repository
• Permission to use any voice or audio you process

1. Download the project files with the hf command.
2. On Windows, double-click run-install.bat. On Linux or macOS, run run-install.sh.
3. Start the app with run-applio.bat on Windows or run-applio.sh on Linux and macOS.

pip install -U huggingface_hub
hf download IAHispano/Applio

run-install.sh

How to use it: Start Applio with its run script, and the Gradio interface opens in your default browser. You convert audio from there. To watch training or visualize data, start TensorBoard with its own script.

run-applio.sh

run-tensorboard.sh

Tool | What it does
run-install.bat and run-install.sh | Install the app on Windows and on Linux or macOS
run-applio.bat and run-applio.sh | Start Applio and open the Gradio interface in your browser
run-tensorboard.bat and run-tensorboard.sh | Start TensorBoard to monitor training or visualize data

Example: You run the start script and the interface opens in your browser. You load a recording you own and convert it. What comes back is the same audio in a different voice. The card shows no screen-by-screen walkthrough, so its documentation is the next stop.

For newcomers and experts: Voice conversion takes a recording of one voice and makes it sound like another. Applio puts that in a web page, so you do not need to write code. The first small thing to try is running the install script, starting the app and looking at the page that opens.

It is based on VoiceConversionWebUI, with a Gradio front end, plugin and configuration hooks, and TensorBoard for training. The card says its official models were trained on public datasets such as VCTK, and it credits gradio-screen-recorder and rvc-cli for parts of the work.

Run the install script
Start the app with its run script
Open the Gradio page in your browser
Convert or train voices
Watch training in TensorBoard

Good to know: Using it is your own responsibility, and the card sets clear terms.

• Terms: by using it you agree to respect copyrights and privacy rights, and to follow the laws where you live.
• Voices: use only audio you own or have explicit permission to use.
• License: MIT. For commercial use, follow it and contact the authors to make sure the use is ethical.
• Cost to run: none named. The card invites optional donations.
• Data it reaches: the audio you process, on your computer. The linked Colab notebooks run on Google's Colab service.
• Not for: harming, defaming or infringing the rights of others, or creating or sharing unauthorized content.

On Hugging Face, 13 November 2026

• Task: audio to audio
• Library: transformers
• Downloads in the last 30 days: 334
• Likes: 229
• License: mit
• Based on: lj1995/VoiceConversionWebUI
• Quantized versions: 1
• Fine-tunes: 6
• Tags: transformers, onnx, vits, voice2voice, audio-to-audio

Links

• Model: https://huggingface.co/IAHispano/Applio
• Files: https://huggingface.co/IAHispano/Applio/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download IAHispano/Applio

Then follow the usage in its model card. Nothing in this kit runs by itself.