OTTO
OTTO copied to clipboard
bitfieldaudio
Finish Phone / Lofi effect
So, since i can't find a good rundown of the effect on video: What is it? It seems like mostly some bitcrushing - perhaps with some granular elements. There's definitely some glitchyness going on...
the OP-1 has an effect that does i think a bit of analogue sounding distortion + definately some bit crushing and some "glitchyness", which seems a bit like a randomly/envelope triggered gate. But i dont think this is necessarilly what we have to do with it. I just think we need something to destroy sound, both analogue-like and digital (bit crushing)
I don't know about the OP-1 effect, but in general, the reason phones sound the way they do is because of heavy signal filtering (e.g. hard low and high passing), so maybe it could be a combination of hard passing, noise modulation, and bitcrushing.
@D-I-S-K-U-S That OP1 effect sounds like Linear Predictive Coding, as used in the Speak&Spell, or Bitspeek for instance. I have some material on that, I'll try and see what I can dig up.
As for the 'traditional' lofi phone sound, yeah, it's mostly highpassing at 300Hz and downsampling at 7kHz. Then, predistortion and bitcrushing can be added for good measure.
LPC is far from trivial for me to implement from scratch, but there are a couple of open source DSP libraries that do it. We could either import it, or inspire from their code.
Soundpipe looks the most interesting IMHO. The only downside is that Soundpipe's lpc is limited to a 10 pole filter lpc. Don't get me wrong, 10 pole is exactly the kinda sound we're looking for, since it's the rough model used in the S&S, but adjustable filter order could have been nice to have.
Either way, if anyone's interested in the maths that are involved, I found those slides to be reasonably exhaustive.
status on this? For me to really be enthusiastic about it, it would need to be not too similar to the OP-1 phone.
Also, @D-I-S-K-U-S We need to rework that UI for the new style. To be honest, I'm not too sure about it anyway, but lets see whats possible in a rework
Do we want to do this before v0.0.1? gives us about two months to have it in
Yeah, we/I should probably come up with a new UI concept for this thats a bit more abstract. On it!
I was just looking through the Faust documentation, and I found there's vocal synthesis. Would that be anything we can work with @nebularnoise? A similar effect at least, not necessarily the same.
Updated. Added GFI flair.
We need to finish this effect. Most of the DSP has already been written, it just needs to be wrapped up in a proper effects engine struct. The code currently in this PR: https://github.com/topisani/OTTO/pull/104 but should be moved to a proper feature/ branch.
Apart from the code, we also need new UI ideas in the new style.