Source-aware generation
Start from a real recording so timing, melody, and musical structure can guide the new interpretation and make the new direction easier to judge against the original.
Upload a source song, describe the new vocal and production direction, and generate a fresh cover with clear prompt and result details.
Review the source song, cover direction, available controls, and account requirements before you begin.
The workflow keeps the source song and cover brief together, so you can test a specific vocal and production direction without losing the musical reference you started from.
Start from a real recording so timing, melody, and musical structure can guide the new interpretation and make the new direction easier to judge against the original.
Control title, style, lyrics direction, negative tags, model choice, and advanced audio weights from one panel. Describe qualities, such as an intimate breathy vocal, rather than requesting a named person.
Generated versions retain task status and playback controls so you can compare outputs before using one downstream.
This focused workflow is more useful than a generic generator when the task, source requirements, and review step are explicit.
The uploaded source supplies musical context that would be tedious to describe from scratch.
Prompt and style fields remain visible, making it easier to revise the brief rather than guess what changed.
Generation creates a new result and leaves the source recording available for comparison.
Prepare a source you are authorized to use, define the vocal and production contrast you want, then compare the generated result with the source before sharing it.
Choose the audio you have permission to use and wait for the workspace to finish preparing it.
Set the desired vocal character, style, title, model, and any traits the result should avoid. Make the contrast concrete, for example from a bright pop delivery to a restrained acoustic interpretation.
Run the task, listen to the returned versions, and refine the direction when another pass is needed.
Use a generated cover as a creative reference or demo, then make separate rights, performer, and final-production decisions before any public release.
Test how a song reads with a different delivery before recording a final vocalist.
Explore a genre or production shift while preserving the recognizable musical foundation.
Produce references for pitches, rehearsals, or editorial planning without replacing final rights review.
A useful cover brief changes specific performance choices while respecting the source, the intended audience, and the rights attached to the song.
Write down what should remain recognizable and what should change. Separate vocal delivery, instrumentation, energy, tempo feel, and production texture instead of asking for a generally different version. A brief such as “keep the intimate verse, widen the chorus, use acoustic percussion, avoid a bright pop mix” gives every result a concrete standard.
Judge tone, phrasing, register, emotional distance, and arrangement rather than whether the output sounds like a named performer. Listen for unstable pronunciation, crowded vocal ranges, and places where the new production fights the original melody. A successful reference should communicate a direction that a real singer or producer could understand and develop.
A compelling demo is not automatically ready to publish. Confirm your permission to transform the recording and composition, review the generated audio for unwanted resemblance or artifacts, and decide whether final vocals should be recorded separately. Keep the source, brief, and selected result together so later collaborators can see how the cover direction was chosen.
Direct answers about source requirements, output, limits, and downstream use.
Move the current result into a relevant generation, editing, separation, or conversion task.
Turn an instrumental into a vocal song by uploading the backing track, describing lyrics and delivery, choosing a style, and generating a new version.
Separate vocals from a stereo song, preview both outputs, and download vocal and instrumental WAV files in a short local workflow.
Continue an existing track from a chosen point by uploading audio, describing the next section, and reviewing the generated continuation.
Bring a permitted source track and a precise creative direction into the workspace.
Create an AI song cover