Applio

Applio is an open-source AI voice conversion suite that provides tools to create voice models, convert voices in real time, and run on desktop or in the cloud. It ships with features for voice conversion, real-time voice changes, model training, and audio processing. The project is community-driven and aims to be auditable, with no telemetry and no accounts required.

The tool supports converting any audio into another voice using ready-made models, with options to tune pitch, clean, upscale, and batch-process files. It also enables live voice changes with low-latency inference, and it allows training a voice from user datasets, as well as blending multiple voice models and generating speech from text. Additionally, an audio analyzer helps prepare datasets and check their readiness before training.

The platform emphasizes openness: it is built as an open-source project under MIT license, runs locally on user machines, and is available for Windows, macOS, and Linux. It provides a desktop app and a Gradio WebUI, with plans to keep the workflow accessible to both beginners and professionals.

Users can access community-driven development, with features, bug fixes, and roadmaps discussed openly. The project highlights no telemetry and no required accounts, focusing on local execution and auditable code. It invites users to download Applio, explore voice models, and test conversions in a playground or via the desktop interface.

Use cases

• Convert vowels and speech in real time or offline for songs, streams, and calls.