My audio lab: how I choose the sound of every song
Every falling-notes video you see starts in what I call my audio lab. Before a song goes out, I listen to it in several versions, side by side, and I choose by ear.
What the lab is
One page, one song, many players. Each player is the same music with one thing changed: the piano, the instrument in the background, or how human the playing feels.
Every version is set to the same loudness. That matters: the ear always prefers the louder one, so if the levels are not equal, you are not really comparing the sound.
The piano comes first
The piano is a real grand piano, recorded key by key at five strengths, from very soft to loud. A soft touch does not just sound quieter, it sounds different, like on a real piano.
I have four colours of the same piano: Accurate, the clear and natural one; Noct, softer and darker, made for quiet pieces; and two more, Twilight and Moonlight. For Now We Are Free I chose Noct.
The human touch
A computer plays every note at the same strength and exactly on time. A pianist does not. So each phrase grows and falls back, the strong beats get a little more weight, and the music breathes a bit at the end of a phrase.
I tried a medium and a strong touch. The strong one won, so every video now gets it.
Choosing the voices
Behind the piano I can now add other voices: strings, an electric piano (Rhodes or Wurlitzer), a harpsichord, and soft drums. They stay in the background, the piano always leads, and only the piano is drawn on screen.
The notes of these voices come from the song itself: the parts that were already in the arrangement. When a song is only piano, the strings follow the left hand, very softly. And they breathe with the piano, so they never drift away from it.
- Same loudness for every version, always
- One change at a time, so I know what I am hearing
- Headphones, then the phone speaker, because that is where you will hear it
- Not every song needs a background: sometimes the piano alone is the most beautiful
Thank you to the people who share their sounds
The piano is the Salamander Grand by Alexander Holm (CC BY 3.0), in the versions remastered by Chisato Yamauchi. The strings are VSCO 2 Community Edition by Versilian Studios (CC0). Free sounds made by generous people: they deserve the credit.
The first song with the new sound is Now We Are Free. Tell me what you feel when you hear it.