Feed live signal to Stretch Player Node

[Oriah Beats]

Hi everyone,
I’m experimenting with the Time Stretch Player and I’d like to use it in an FX plugin context for vocals. My idea is to take the incoming vocal signal in real-time and then process it through the Stretch Player to make it faster (or slower) so I have 2 questions

is it possible to feed live input audio directly into the Stretch Player, or does it only work with preloaded samples?

Are there any performance considerations (CPU, latency) I should keep in mind if I want this to run live? I’d really appreciate any guidance or examples from anyone who has tried something similar

[HISEnberg]

@Oriah-Beats it only works with a preloaded sample, you will have to create your own solution with either Faust, Snex, RNBO, or C++ to accomplish this on a live signal.

[DanH]

@Oriah-Beats someone got rubberband working in Hise recently...

[HISEnberg]

@DanH Nice good call that is an elegant solution!

[Oriah Beats]

@HISEnberg I understand. my knowledge is limited to create my own Faust. have you done something similar before?

[Oriah Beats]

@DanH is this the thread ? https://forum.hise.audio//post/108326

[DanH]

@Oriah-Beats yup!

[HISEnberg]

@Oriah-Beats I've experimented with it but I've never fully implemented it.

There's a few different methods to accomplish it, but I think the simplest (buffer and resample) is fairly simple to accomplish. You can also look into granular synthesis which is fundamentally pretty similar.

I would also recommend looking at building a phase vocoder which would be the best in terms of audio quality, but this is a more complex topic.

[Oriah Beats]

@DanH I will give it a shot

[Oriah Beats]

@HISEnberg I am abit lost but let me do more research and see if I can implement this