Looking through the code, I can see that you wrote some code to premix the files into a single "track". This is cool and it never occurred to me that this approach can work well (which it does clearly!).
I just have a general curiosity/architecture question for you?
Why did you choose to premix in the first place versus playing the individual samples in real time? If they're managed individually you could potentially build a UI/Command-line component where you can view the sequencer and turn on or off each instrument per note?
Really cool lib btw!
-deckarep
Looking through the code, I can see that you wrote some code to premix the files into a single "track". This is cool and it never occurred to me that this approach can work well (which it does clearly!).
I just have a general curiosity/architecture question for you?
Why did you choose to premix in the first place versus playing the individual samples in real time? If they're managed individually you could potentially build a UI/Command-line component where you can view the sequencer and turn on or off each instrument per note?
Really cool lib btw!
-deckarep