Application for transcription of vocal voice to music notes.
The application was created as a result of a master's thesis - Algirithmization of vocal voice transcription to musical notation. The following version of the application is a reduced version due to the fact that it does not support one of the most important libraries on Windows.
This version has the functions that allowed to get the best transcription results of all that were tested and created during the work.
Have you ever imagined a great guitar solo, but you couldn't write it down?
You know how to sing it and you know you'll forget it in a minute....
Now you can sing short melodies, the notation of which you will receive.
(-1. Recording the melody)
(0. Install Lilypond and library abjad if you want to get score and play your result melody)
- Select the segmentation method (for now, it is automatic in total)
- Select file (select what you want to analyse (the wave file you recorded earlier))
- You will receive notes in the form of text (score or recording)
* Make sure that only your voice is on the recording
* Accent the beginning of the notes clearly
* Sing precisely
* and preferably do not use the text - sing on the syllable of your choice.
* Segmentation of recording (where a new note begins)
* Frequency analysis (what is its pitch)
* Musical analysis (what musical pitch it has and how to write it all down)
* Creating notes
* Display and listen (if you have an abjad)
In current programme:
* Displaying notes in the graphical interface
* Completing graphical segmentation
* Ability to edit notes and re-display
* Tonality control
Maybe in the future, its optional inspiration:
* Pause detection
* Advanced music analysis (e.g. based on Markov Models)
* Suggesting melody direction
* Text recognition
* Transition to object-oriented programming
* Graphical note processing