You point it at a folder
Local files, on your own disk. Mixterio listens to the audio and ignores the tags, so a wrong key or BPM written by some other tool cannot poison the set.
One folder is enough. The free tier is built around exactly that.
Methodfolder in, night out
Six steps, and none of them are guesses.
Most tools decide a transition in the moment it happens. Mixterio does the work early: it builds the blend as audio, measures it, and only then lets it into a set. Here is the whole thing, in order.
Seam study, 32 beatsgrid lock
24 to 40 beats The classic. The lows change hands on a phrase boundary.
Positionbeat 00 / 32
Tempolocked, 0.00%
Low bandoutgoing
Gatemeasuring
A diagram, drawn from the engine's own thresholds and its own beat ranges. It is not a recording and not a screenshot. The real thing is built as audio, measured, and only then allowed into a set.
It studies every track once
Beats and downbeats, tempo and how steady it is, key, loudness, the energy across the track, where the intro and the drop and the outro sit, and the drums, bass, vocals and melody pulled apart. It keeps all of that, so it never has to listen twice.
This is the slow part, and the reason everything after it is exact.
It builds transitions before anyone hears them
For a pair of tracks that might work, the engine builds the actual blend as audio: levels matched, the bass handed over on a phrase boundary, the incoming track arriving exactly on the beat. Real audio, not a prediction.
Idle machine time plugged into power goes here. So does every transition you play.
It grades what it built
Then it measures the result against fixed standards: does the tempo hold, does the incoming track land in time, does the low end stay clean, does the new track properly arrive instead of appearing out of nowhere. Pairings that pass are the ones it will use. The ones that fail are remembered as failures.
The standard does not move. Only how much of your library has cleared it.
It composes, edits and exports
It picks the sequence at one locked tempo, chooses how each pair should be joined, and hands you an editable draft. Swap or remove tracks in the Set Workbench, export directly to Rekordbox XML with cue points or an m3u8, or take the tracklist.
You can stop here, export to CDJs, or take the tracklist.
It performs, and keeps learning
It plays from the audio it already built rather than deciding in the moment, comes back from a crash where the room stopped, and takes direction in plain language. Every transition it plays gets measured and remembered.
The background supervisor keeps measuring while idle, which is why the app is better in month two than in week one.
The thresholds
What a transition has to clear to be allowed out.
These are the engine's own numbers, not marketing ones. They are the reason a set can be left alone.
Measured-map floor0.85
A pairing only counts as proven once its transition has been built as audio and scored at or above this. If the engine ever has to reach below the floor, it tells you on screen instead of playing it quietly.
Established handoff3.0 dB
The incoming track has to arrive into a section that is already running. If it would jump in louder than this, the blend is stretched to the next phrase or a different track is chosen.
Tempo lock6%
Tracks too far from the tempo of the night are ruled out before anything else is considered, so nothing gets stretched into a chipmunk.
Legal warp band12%
A hard ceiling on how far any track can be pulled from its own natural tempo, however much the rest of the set would like it to fit.
The difference
The same pair of tracks, two ways.
One of these paths finds out in the room. The other finds out on your machine, hours earlier, where a failure costs nothing.
The usual way
- Compare key, tempo, energy
- Suggest a cue point
- Apply a curve live
- Hear it for the first time in the room
Our way
- Build the blend as real audio
- Grade it: timing, tempo, low end, arrival
- Below the floor, it is refused
- The room only ever hears a pass
What this costs, honestly
The trade we make on purpose.
Because Mixterio studies a track before it can mix it, you analyse your library first. That is a one-time pass per file, and until it is done a track cannot be used. There is no instant mix of something the engine has never heard.
It also works from your own files. No streaming catalogue, no cloud library, nothing that depends on a venue's network at midnight. Your music and your analysis stay on your machine.
Both of those are the price of a transition that was graded before it played. If you want something that mixes anything instantly with no preparation, that exists, and it is not this.
In exchange
- 01
Every transition was measured
Not guessed from key and tempo. Built as audio, then graded.
- 02
The night is stable
One tempo, no repeats, and a resume path if the machine falls over.
- 03
It gets better where it sits
Idle machine time and ordinary use both add proven pairings from your own collection.
Want to be told when it ships?
One email at launch. Nothing else, and nothing to buy today.