MPC Vocoder or Robot Voice?

I have been connecting a mic to the inputs of the MPC Keg 37 and trying to create something like a vocoder or robot voice effect. The best results I've had came with the pitch shifter. Is there any way to create a real vocoder? What are your favorite modulations for voices? Thanks

> MPC Keg 37
> Robot Voice

MPC Keg 37


I saw that earlier and 'd

Maybe talkbox effect first, have LFO work Formant and other Params, then through rhe band pass filter while automation modulating params with LFO, rhem through a filter gate or the trance gates in Flexbeat? Migjt need to granulate and/or bit crush before BP filter.

Lampdog wrote:MPC Keg 37


I saw that earlier and 'd

Lampdog wrote:I mean some people make beats while drinking and drink while making beats, that machine could be a win/win.

That Beer Beats Mixtape is coming soon!

Dennis.Freeman wrote:I have been connecting a mic to the inputs of the MPC Keg 37 and trying to create something like a vocoder or robot voice effect. The best results I've had came with the pitch shifter. Is there any way to create a real vocoder?

The way a vocoder works I don't think you can create one with built in FX. You need multiple bandpass filter with envelop followers on the input signal, and the same number of BP filters controlled by envelop followers.

What comes close to it is the Multiband filterbank. But it would need to have a built-in synth for the actual output. And of course more bands. With just 3 bands the voice would be unintelligible.

I wouldn't be surprised if the number of vocoder FX for a project would be 1 (for gen1 hardware).

Masi wrote:

Dennis.Freeman wrote:I have been connecting a mic to the inputs of the MPC Keg 37 and trying to create something like a vocoder or robot voice effect. The best results I've had came with the pitch shifter. Is there any way to create a real vocoder?

The way a vocoder works I don't think you can create one with built in FX. You need multiple bandpass filter with envelop followers on the input signal, and the same number of BP filters controlled by envelop followers.

What comes close to it is the Multiband filterbank. But it would need to have a built-in synth for the actual output. And of course more bands. With just 3 bands the voice would be unintelligible.

I wouldn't be surprised if the number of vocoder FX for a project would be 1 (for gen1 hardware).

“Maybe talkbox effect first, have LFO work Formant and other Params, then through rhe band pass filter while automation modulating params with LFO, rhem through a filter gate or the trance gates in Flexbeat? Migjt need to granulate and/or bit crush before BP filter.”

So this post made me try something for the first time. I have an app in my iPhone/iPad called Voice Synth. It is an incredibly deep vocoder/pitch corrector and more which I never had a real use case for. I have a Live 3 (although I will test on my 2 which is Keg37 gen, when I get a chance). It literally took less than a minute to load some vocals on a track, send to USB1-2 and create an audio return from USB 1-2 on a new track and I had my vocoder up and running. Some of the presets are outstanding, but the level of control is as good or better than a top shelf vocoder VST. https://voicesynth.com

The other option I want to try is to just use the iPhone as a microphone and only sample the Vocoded sound from Voice Synth through USB-C. More restricting, but in some cases might be perfect.

I haven't tested the midi control yet for harmonizing notes, but I have previously with some of my synth plug-ins which work great. FYI, the official Moog iOS plugins are a steal and to my ears sound better than the model D MPC plugin.

Hope this helps someone, I am stoked I have a totally new workflow. Thanks for this question. I watched the thread for a few days looking for answers until I got off my ass and did the experiments.

.a