We are proud to be able to announce the development of an audio recognition app compatible with the desktop version of ScoreCleaner. Release date is as of yet TBD but the app enables anyone who can whistle or sing a tune to have it rendered into correct notation and MIDI, synch it with ScoreCleaner and extend it into a full score using ScoreCleaner or any other notation tool of choice.
A message from our CTO Sven Emtell:
As you probably know, the input into ScoreCleaner right now is MIDI information. Either played directly or as Standard MIDI Files containing performance MIDI or quantized MIDI.
For some time now we’ve been working on the next leap, where the input into ScoreCleaner is audio. I use the term leap instead of step, since moving from MIDI input to audio input is not trivial. It involves both signal processing and musical knowledge for the result to be usable.
We’ve been researching audio input and what might be an acceptable level of accuracy when it comes to interpreting audio input to notation and MIDI and we’re proud to announce that we’ve got proof of concept of an audio recognition solution that is a great start on the road to what we would consider one of the best audio recognition solutions on the planet.
Using our, as we like to think, unique knowledge of musical structure, we are convinced that this can really make the foundation of a great product. Take a look at the video below and stay tuned to get updates on further development.
Sven Emtell, DoReMir Music Research
ScoreCleaner Notes Video