Choose audio
A future VocalSep workflow is expected to begin with a user-provided audio track. Supported formats, file limits, and availability will be published when confirmed.
Separate the mix
An AI model analyzes the recording and estimates target sources, such as vocals and accompaniment. Separation is an estimate: artifacts or bleed between parts can occur.
Review your parts
Listen to the resulting audio and decide how it fits your project. Do not assume any particular export format or processing time until product details are announced.
Use audio responsibly. Only upload or reuse material you have the right to process. A separated track does not remove copyright or licensing obligations.