Music transcription has been around for decades, algorithm-wise. In principle this one should actually be easier since it doesn't need to transcribe from scratc

Music transcription has been around for decades, algorithm-wise. In principle this one should actually be easier since it doesn't need to transcribe from scratch but "only" find where in a given sheet you are most likely to currently be. (Next step: evaluate afterwords and point out mistakes)