Your script goes in — a game level, a ten-minute video, a documentary chapter. We read it, cut it into the moments that matter, and give each moment its own cue, plus a voice-over on every block if you want one. No click at the loop point, no track stretched across ten minutes, no loudness fight with your sound effects. Every file arrives loop-ready at −16 LUFS, stems included.
Runs on our real demo cues. Same reading, same files — this is the product working, not a mockup.
ENGLISH · PORTUGUÊS · ESPAÑOL — the reader routes by meaning, not by language. Nothing is uploaded: the breakdown happens in your browser.
Most people don’t fail because they have no music. They fail because they have to choose one track for a video that has twelve different feelings in it — and they don’t have the vocabulary to describe any of them.
Script, caption, storyboard, level description, the transcript of your video. Any language. You don’t have to name a genre or a tempo — that’s our job.
Sentence by sentence, the text is cut into beats: this is the opening, this is the build, this is the hit, this is the aftermath, this is the resolve.
Each beat gets music that fits it — the same key and tempo family, so the cuts don’t jump. Turn the voice on and every block also gets its narration, timed to the same map.
WAV with the loop point written inside, OGG for the engine, stems for the mix, everything at −16 LUFS so your sound effects still breathe. Drop in, play, ship.
Each one costs a weekend to find, a day to understand and a night to fix. They are already fixed in the files you download here.
Every music file ends. In your engine, that ending is a silence followed by a jump — an audible click on every restart, or the music dropping out mid-fight. Nothing in a normal download tells the engine where the loop begins and ends.
Music made for Spotify arrives at −9 LUFS because it is competing with other songs. Put that under your sound effects and the footsteps, the gunshots and the doors vanish. You then spend a night turning everything up until it clips.
The visitor arrives, the fight starts, the ending lands — and the music is still the same smiling loop from minute one. To fix it you either crossfade in an editor you don’t have, or you buy twelve tracks that don’t share a key.
Same product, six jobs. Every cue below was measured after rendering: tempo, brightness, low end and transient density checked against what that kind of scene actually needs.
All six: 60 s · instrumental · −16 LUFS · seamless loop · stems on paid plans.
Turn the voice on and each block of your text is spoken — in the language of the text — with the music already ducked underneath it. You don’t mix anything: the background drops to −32 dBFS under the voice and comes back in the gaps, so nothing competes.
The voice sits 15 dB above the music, with a −24 dB cut in the 1.5–4 kHz band so it never fights the consonants. It sounds like a finished edit, not a robot reading over a track.
No monthly plan, no per-video license. You get the files, you keep them, and you can use them in commercial work.