The sounds I’m making when training Dubler’s triggers are different to the sounds I’m making mid stream. This means Dubler is having issues recognising my triggers. The problem is that I don’t have total awareness of the sounds I’m making mid stream.
I thought it’d be cool if I was able to record a longer waveform of me beatboxing over like 16 bars in Dubler and then select portions of that waveform to train Dubler. Like “hey dubler, these are my snares, train with these portions.”
That way Dubler could be trained with the snares I’m actually using, rather than the snares I think I’m using.
This is a really cool idea, and something we’ve also thought about here at Vochlea. We’ve got a few ideas for trigger recognition that we’re working on. We’ll update users as we make progress on them and what we’ve got coming up in future updates