You are here

Acoustic book

The encoding MIDI

The MIDI protocol (1982) is a very efficient method of representing the information related to music performance.
This makes it very interesting also for the sound management in computer applications, such as multimedia systems or games.
MIDI files do not contain audio data, but only execution instructions for the synthesizer that must produce sound stimuli. These instructions are in the form of MIDI messages that indicate to next equipment that notes to play, with what intensity, for how long and on what instrument.

MIDI messages contain executive information, including:

  • Note On (instant when a certain note is activated + the number of the corresponding button)
  • Note Off (instant when a certain note is deactivated + number of the corresponding button)
  • Velocity (force with which the button is pressed: usually it controls the volume)
  • Program (it specifies the type of instrument to play)
  • Time and relative changes (slowing down or speeding up)

The MIDI data are then sent to a sound module, or Expander, that responds to messages producing sounds already stored with different techniques. In this way, you have a huge memory savings (at least three orders of magnitude) compared to the storage of audio encodings. With a good Midi encoding and with a good Expander very good results can be obtained.

With some sequencers is also possible to mix MIDI tracks and audio recordings.

From a MIDI file you can also get the encoding of music scores. This encoding usually requires that time data are previously quantized, that is aligned with the typical values ​of music figures, in a way to produce a correct score, taking into no account the executive time imperfections.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer