The sound recording must have a sample rate of 44100 Hz
The underlying physical model is generic enough to recognize complex tones of virtually any pitched musical instrument
Starting with single notes, there is effectively no limit to the number of notes that are recognized simulteneously
The advanced settings allow you to tune the algorithm to your instrument. You can even train your own model
The algorithm is compact and fast. Even without fancy optimization, the algorithm runs faster than real time
An account at Algoriffix is totally free, yet it will grant you several benefits!
The music is not in the notes, but in the silence between.
If I were not a physicist, I would probably be a musician.
When you play, never mind who listens to you.
I would rather write 10,000 notes than a single letter of the alphabet.