Section of the musical writing system workflow.

Scientific approach

Musical Writing System is an algorithm by Glauco Leo filed at the italian patent office UIBM with n.102018000004364, called Method for the representation of a sequence of sounds in a textual form. It is an extension for audio analysis softwares which could be used together with spectrogram and waveform.


This invention comprises a method for the representation of a sequence of sounds in a textual form which allows to identify and analyse them. It comprises, but is not restricted to, biology and music field.
The invention allows to describe sounds which have frequencies between 0 and 150.000 Hz by an accuracy of their durations to within ± 0,0078125 and combinations of a limited number of typographic symbols. It only covers frequency (F) and duration (D) of sounds and well-developed combinations of symbols are linked to them. The combinations are based on default elements and thanks to their connections it is possible to generate infinite combinations starting from a finite set of symbols. F symbols are 9 and they generate an At finite sequence of 25 combinations you can replicate indefinitely, of which Ii represents the default unit of measurement and it has a duration of one second. D symbols are 4 and you can use them to generate an infinite sequence of Cs combinations. To make the process easy a default M set of 128 Cs combinations has been generated, from which it is possible to get infinite Cs combinations. Each At and Cs value is linked to an approximate value of X for frequency and V for duration respectively. Xy is an infinite sequence of logarithmic scale values, the basis of which is the first useful F value, and Vs is an infinite sequence of duration discrete values expressed in seconds. By two functions, the closest X value to F and the closest V to D are assigned. So durations can be represented by an accuracy of ± 0,0078125 which allows the main goal of the method to be get, or encoding to be comprehensible to a human being, and not only to a calculator, without compromising milliseconds description. There are 4 markers which complete the typographic symbols set of this method, and their use is common to all the elements of the encoding. The relationship between such combinations gives the final result of the encoding. In short, the process is the following. Given an input Sn sound you can get the corresponding Fn and Dn values. You can link to Fn, by ƒ(F), a Xy value from which you can get the corresponding At combination of symbols. The result is the encoding value of Fn, or Tn=XyAt. Then the corresponding Dn of Sn is read. You can link to Dn, by ƒ(D), a discrete duration Vs value from which you can get the corresponding Cs combination of symbols. The result is the encoding value of Dn, or Qn=VsCs. The encoding value of Sn is Rn=AtCs , and it is only composed of values represented by typographic symbols, or by the encoding elements which are useful to represent the result in a textual form. The process is repeated until all the Sn are encoded, then it is completed. The result of the encoding is the transcription of an audio composed of a sequence of sounds into a textual form, comprehensible to a human being and a calculator at the same time.
This method allows either the transcoding of a digitized live-recording audio signal or the transcoding of a pre-existing digital audio. In biological field, or ornithological, but shall not be limited to, it is possible to comprise a system to import a sequence of sounds you can transcode by a specifically developed device, it to be make useful for data transmitting and printing to video of results, with the possibility to identify the source which emitted that sequence by comparing it with pre-existing files placed on a local or remote database.
As it now stands this method is limited to the encoding of a signal composed of a sequence of sounds, but it could be easy to develop it for simultaneous sounds transcoding.

musical writing system logo audiografo logo