Suno AI Music

Describe a scene, a feeling or a genre and the model writes lyrics, melody and arrangement in one pass. Custom mode gives you separate fields for lyrics, style and title, so you can keep your own words and let Suno handle everything else. Each run returns two takes, which is useful because the first one often misses.

Suno Studio is a browser-based DAW built around generated audio. It handles arranging, MIDI, effects, automation and design-your-own plugins, so a track can be finished without exporting to Ableton or Logic. Studio access sits on the Premier plan, and it's the main reason heavy users pick that tier over Pro.

Record a few seconds of your own voice and the model sings anything you write in that timbre. A verification step asks you to read a random phrase so a cloned voice can't be created from someone else's clip. This is the feature that moves output from "a generated song" to "my song" for singers who don't like their own takes.

Paid users can train a custom AI model on their own recordings, so new tracks start from their sound rather than a generic average. It suits producers with a catalog who want consistent texture across a project. The tradeoff is time spent training and a narrower range when you want to go somewhere unfamiliar.

Stem separation pulls a finished track into time-aligned parts: vocals, drums, bass and more, depending on your plan. Those stems drop into any DAW for a real mix. Remix goes the other way, letting you change the style of an existing song, rewrite its lyrics or rebuild it around a new recording.

Style and weirdness sliders, exclusions, vocal gender and inspo references give you steering beyond a single sentence. Small moves change a lot, so the practical approach is to lock the lyrics and nudge one control at a time. It's the difference between a random result and a repeatable one.