Using the Sample Editor’s “Audio to Score”

You can use “Audio to Score” to transform a monophonic audio recording into musical notation by creating a MIDI region that corresponds to the recorded melody.

Note: Good results can be achieved using clearly identifiable, monophonic audio material. This function is best-suited to producing melody notes from a clearly sung non-legato vocal line. It works best on unprocessed audio—that is, audio without effects.

Opening the Sample Editor’s “Audio to Score”

You need to select an area within an audio file before you can open “Audio to Score.”

To open “Audio to Score”
  1. Select the destination MIDI or software instrument track in the Arrange area.

    The MIDI region generated by the “Audio to Score” function will be placed on this track.

  2. Select the area of the audio region in the Sample Editor that you want to process through “Audio to Score.”

  3. Choose Factory > “Audio to Score” (or use the corresponding key command, default assignment:  Control-S).

    Figure. "Audio to Score" pane.

Setting “Audio to Score” Parameters

Here are the parameters in Audio to Score. You may need to experiment with different parameter values, and even make several attempts to determine the optimum settings for a particular audio recording.

Figure. "Audio to Score" pane.
  • Granulation: Determines the time span of louder components in the audio material. Logic Express uses these peak signals (or transients) to discriminate between notes that it should—or should not—analyze. The most useful values are usually between 50 and 200 ms, depending on the tempo of the audio material.
  • Attack Range: Determines a suitable attack phase length for the audio material. For example, drum and percussion instruments have short attack times (less than 20 ms), whereas string instruments have longer attack phases. The best values for most instruments are usually between 5 and 40 ms, with most around 20 ms.
  • Smooth Release: Specifically designed to process audio material that contains sounds with a long release or reverb tail. This makes it easier to convert these sounds to notes. The value you choose here should generally be between 0% and 5%, except when you are processing passages that contain sustained notes, distorted guitars, or similar sounds.

    Note: The quality of “Audio to Score” results will benefit from non-legato performances. Therefore, you should try to avoid reverb or release tails in your audio material, or at least keep them to a minimum.

  • Velocity Threshold: Sets the threshold level. All signals that fall below this value are ignored. In most cases, you should choose a value of 1, except when processing very dense, loud material with soft background noises.
  • Time Correction: Compensates for any time delays that may occur when external samplers or synthesizers are triggered by MIDI notes. These time delays are sometimes very noticeable, particularly if the connected device is playing a MIDI region that was generated by the “Audio to Score” function alongside the original audio material. You should be able to compensate for this effect by using settings between −20 ms and 0 ms.