A sentence becomes a finished track in about thirty seconds — vocals or instrumental, with stems, BPM and key attached. No instrument, no session, no plugins. Just the thing you were already humming.
3 FREE TRACKSNO CARDMP3 + STEMS
18,400 tracks written here this month
Let the Rain Do the Talking
@lune
Paper Light
@lune
Slow Tide
@rue
Third Floor, 4AM
@kaito
Window Seat
@juno
Long Exposure
@lune
Black Glass
@nova
Undertow
@kaito
Red Shift
@nova
Hollow Engine
@kaito
Night Shift
@sable
Static Bloom
@rue
Sunroof Down
@orbit
Ninety Nine
@orbit
Confetti
@juno
Lemonade Static
@juno
Let the Rain Do the Talking
@lune
Paper Light
@lune
Slow Tide
@rue
Third Floor, 4AM
@kaito
Window Seat
@juno
Long Exposure
@lune
Black Glass
@nova
Undertow
@kaito
Red Shift
@nova
Hollow Engine
@kaito
Night Shift
@sable
Static Bloom
@rue
Sunroof Down
@orbit
Ninety Nine
@orbit
Confetti
@juno
Lemonade Static
@juno
Afterparty Lights
@orbit
Pace Yourself
@kaito
The Last Light
@sable
Cold Open
@sable
Glacier Choir
@lune
Dust Road
@sable
Ember Theory
@nova
Cosmic Descent
@mehdi
Ancestral Kick
@mehdi
Behind the Lie
@mehdi
Into the Void
@mehdi
Hall Anthem
@mehdi
Wave of Gold
@mehdi
Time and Energy
@mehdi
Futur Soul
@mehdi
Maybe
@mehdi
Afterparty Lights
@orbit
Pace Yourself
@kaito
The Last Light
@sable
Cold Open
@sable
Glacier Choir
@lune
Dust Road
@sable
Ember Theory
@nova
Cosmic Descent
@mehdi
Ancestral Kick
@mehdi
Behind the Lie
@mehdi
Into the Void
@mehdi
Hall Anthem
@mehdi
Wave of Gold
@mehdi
Time and Energy
@mehdi
Futur Soul
@mehdi
Maybe
@mehdi
Made on Wavvy this week
Hear all of itHow it works
Wavvy is a text box and a player. Everything between them is handled.
A sentence is enough. Stuck? The composer rewrites your line into a proper brief — style, instruments, tempo — before anything is spent.
Around thirty seconds for a full take, with BPM, key and the arrangement marked on the waveform. Generate again and the earlier takes stay put.
Export MP3, pull the vocals off the instrumental, or publish to the feed where anyone can remix your prompt into their own version.
The studio
Most tools hand you one track and clear the screen. Wavvy keeps the whole session in front of you, with the recipe attached to every take.
Write it however it comes out. One tap turns it into a brief a producer would recognise — and writes the lyrics too, if you asked for vocals.
sad song for the rain
Downtempo neo-soul ballad, 68 BPM in F minor. Rhodes piano, brushed drums, upright bass, soft female vocal with a close, breathy delivery.
Every take stacks in the session panel with its prompt, seed, BPM and key. Regenerate freely: take #1 is still there when take #7 misses.
Rain Talking
Rain Talking
Split vocals from the instrumental, download MP3, and carry the key and tempo straight into your DAW. Half a generation per split, no subscription gate.
Publish a track and its prompt travels with it. Anyone can open it, remix the wording, and post their own version — credit intact.
Remixed from @nova — “black glass, slowed”
+412Advanced mode
One prompt is a good start. When you know the shape you want, switch to advanced and build the track section by section.
Order the sections. Intro, verse, chorus, bridge, outro — add, reorder and delete them until the shape is right.
Style each one separately. The chorus can lift while the verse stays nostalgic. Styles are set per section, not per track.
Set the length in seconds. Each block carries its own duration, and the total is what you are charged for — before you spend it.
Lyrics where they belong. Write or generate lyrics on the sections that sing, and leave the instrumental ones alone.
On the timeline
Comments pin to a moment, not to the bottom of a page. Someone hears the drop land wrong at 1:04 and that is exactly where their note sits.
Let the Rain Do the Talking
BPM 70KEY Bb
Click any face to jump to the second it talks about.
3 comments on the timeline
Notifications tell you who commented and where. Reply from the track page — the timestamp is part of the reply.
Made for how you work
Wavvy never asks who you are. It reads the track — tempo, key, arrangement, whether there is a voice on it — and proposes the workflow that track is asking for. Eight professional contexts, one click apart, because you are a DJ on this one and a composer on the next.
A beat grid, cue points, 32 bars to mix into — and a set that already knows what follows what.
32 bars on both ends
Ask for an intro and an outro and the track comes back with room to mix in and out of.
Acapellas that exist
Isolate a clean vocal and drop it over anything else in your crate.
The set orders itself
Wavvy knows which of your tracks sit within 6 BPM and a compatible key. It proposes the order.
Edits
Prepare
Perform
Wavvy proposes this from the track itself. Nothing is remembered — the next track gets its own read.
What happens after generation
Generating a song is the cheap step, and everyone can do it. What decides whether a track ever gets used is what comes next — and that is the part Wavvy is built around.
Ancestral Kick · 122 BPM · Dm
live outputReads as a fit for the DJ workflow — 122 BPM, Afro house, built around a drop.
Tempo, key and section boundaries are measured from the file itself; the workflow is proposed from them.
Pricing
Credits buy generation time — 900 a minute. Unused credits roll into next month, and you can stop any time.
$0/mo
3 tracks a month
See whether it writes what you hear.
$6/mo
~33 min of music a month
For writing more than you keep.
$22/mo
~134 min of music a month
For everything that goes out to an audience.
$48/mo
~300 min of music a month
For the people who finish it elsewhere.
Every plan publishes to the feed, keeps timestamped comments, and lets anyone remix your prompt. Credits buy generation time at 900 a minute; stems and edits cost half a generation. Change or stop a plan whenever — unused credits roll over.
FAQ
Still deciding? Open any track in the feed and read the prompt that made it — the whole catalogue is public.
Browse the feedYou do. Everything you generate is yours to keep, post and edit. Commercial release is covered on the Creator plan; on Free and Starter the tracks are yours for personal and social use.
No. If you can describe a scene, you can brief a track. The composer fills in tempo, key and instrumentation, and shows you what it chose so you can push back on it.
The ElevenLabs Music API, called server-side. Your prompt goes out, the audio comes back, and the key never touches the browser.
Yes. Split a track into vocals and instrumental, download it, and the BPM and key are on the card so it drops straight into a session.
It travels with the track. Anyone can read it and remix it — that is the point of the feed. Keep a track private if you would rather not share the recipe.
3 free tracks · no card · cancel whenever