What we tested in the current converter
We generated a 22,050 Hz mono WAV containing the same 440 Hz A4 tone three times. Each tone lasted 1.2 seconds, with 0.35 seconds of silence between tones. The sample amplitudes were 0.18, 0.48, and 0.86. No video, performer, title, channel name, or copyrighted recording was used.
| Generated tone | Sample amplitude | Detected MIDI note | Note On velocity |
|---|---|---|---|
| First | 0.18 | A4 | 64 |
| Second | 0.48 | A4 | 72 |
| Third | 0.86 | A4 | 81 |
The browser produced a 78-byte MIDI file with three notes and no console errors. Their start times were approximately 0.000, 1.532, and 3.067 seconds. The rising velocities followed the rising amplitudes in this file, but the three numbers are model estimates rather than calibrated sound-pressure or decibel readings.
Velocity is not the same as audio volume
MIDI velocity is data attached to a note event. A software instrument may use it to change loudness, attack, brightness, or sample choice. The same velocity can sound different with another instrument or preset. MIDI 1.0 Note On messages use a 7-bit velocity field; the MIDI Association's message list documents the 0–127 data range.
The exported file does not contain the source waveform, a decibel curve, microphone gain, room sound, compression, or the original mix. It also does not create continuous volume automation. It contains note instructions, including the note number, timing, duration, and an estimated Note On velocity.
Why a loud recording may not make every MIDI note loud
Peak sample amplitude and perceived loudness are different things. Compression can make a recording sound consistently loud while reducing its peaks. Reverb can extend energy after a note. Several instruments can overlap, and the converter may attribute energy to the wrong pitch or combine it into one note estimate. That is why the exported velocity should be reviewed alongside pitch and timing.
How to fix dynamics in a DAW
- Import the MIDI onto an instrument track. Choose the sound before judging the dynamics, because the instrument determines how velocity is interpreted.
- Open the velocity lane. Look for sudden values that do not match the musical phrase.
- Edit the shape, not only individual spikes. Raise or lower groups of notes to restore crescendos, accents, or a stable accompaniment.
- Use track volume for the overall level. Velocity changes note expression; the channel fader changes the track's output level.
What this test does and does not prove
It verifies that the current local-file converter can write different Note On velocity values and that a simple rising-amplitude input produced a rising sequence in this run. It does not prove that every recording preserves dynamics, that velocity is linear with amplitude, or that the values will sound equally spaced through every synthesizer.
If the draft has unwanted short detections, use the extra-note cleanup test. If it plays with the wrong sound, read the instrument and playback guide. For mixed recordings, check the track-separation boundary.