The music API with rights & provenance as first-class objects.
Generate songs, music, vocals, and stems from one endpoint catalog -- then see exactly which license applies, what each call cost, and which key spent it. All in one console.
By getting a key you agree to our Terms of Service.
Capabilities
Every capability maps one-to-one to a documentation page with real, current pricing.
Prompt-to-song with vocals, across v5, v6, and the v7 flagship.
Instrumental generation across genres, tempos, and moods.
Structured, rhyme-aware lyric generation and co-writing.
Turn lyrics into expressive sung vocals.
Continue or lengthen a track -- intros, outros, full arrangements.
Re-imagine a reference into a new style while keeping the hook.
Create a reusable private voice clone from a consented reference.
Rewrite a selected section while preserving the rest of the song.
Split a track into vocals, drums, bass, and more.
Speech-to-text transcription for vocals and dialogue.
Generate sound effects from a text description.
Apply production effects to an existing audio file.
Check a track for copyright matches before you ship it.
Quickstart
From key to first song in three calls.
Create a key on the API Keys page. Pass it as a bearer token.
POST your prompt to a generate endpoint. You get a message_id back.
Poll GET /v1/generations/{task_id} until complete, then fetch the asset.
Open the full Quickstart in Docs →
Trust & provenance
The differentiators enterprise buyers ask about.
Models trained on licensed and consented data.
Voices and styles are opt-in and revocable.
Every call's cost and license tier is visible in your ledger.
Royalty-Free through Full Ownership -- upgrade rights per track, self-serve.