Virtual Instrument routing as a Vocoder

[tiagodb]

 

Hi everyone,

I see many Ableton Live users creating Vocoder-type voices through a timbre coming from some virtual synth like Serum, Diva, Sylenth1, you name it. That is, the Vocoder modulation doesn't come directly from the "Vocoder" instrument, but rather from the selected timbre in a virtual synth.

 

I need to do this in Logic. Is it possible? What's the best solution for this?

Thank you very much.

 

[des99]

Sure, Logic has a vocoder in a few forms, and you can do this using the synth instrument version (EVOC 20 Vocoder Synth), and feeding the modulator (your voice) in via the sidechain (called the "Analysis Input" in that instrument).

[tiagodb]

2 hours ago, des99 said:

Sure, Logic has a vocoder in a few forms, and you can do this using the synth instrument version (EVOC 20 Vocoder Synth), and feeding the modulator (your voice) in via the sidechain (called the "Analysis Input" in that instrument).

Ok, that works but it doesn't address the detail of me wanting to use synthesis with a virtual instrument in the mix. I don't want to use just a Vocoder plugin; instead, I want a Vocoder plugin that allows the use of a synth like Serum to then do the modulation.

[des99]

Again, you can use Logic's vocoder for this (it comes in three versions, check the manual for how to route audio etc), or you can buy a third-party plugin instead.

[Dr.Socrates]

I am quite possibly wrong here, because I know that des99 is very often right, but I think that there is some misunderstanding going on here, and that it is in fact not possible to do what tiagodb asks for using the EVOC. Please correct me if I am wrong!

First, some terminology, because it is a bit difficult to understand what you are asking for, tiagodb. As per the Logic manual, a vocoder basically has two main ingredients, an “analysis signal” and a “synthesis signal”. In EVOC, you can e.g. use an audio input, a pre-recorded audio, the output of some other track or a bus as the analysis signal, which is routed in to EVOC using the side-chain setting, as des explained already. So, des is correct that you can use your own voice, or the output of some other synth (Logic's own or 3rd party) as the analysis signal. Or a drum beat. Or whatever sound you like. But, if I am not mistaken, EVOC always uses its own internal synth for the synthesis signal. 

Now, I think that what tiagodb asks for is to be able to use one audio source (e.g. a voice) as the analysis signal, but another external sound source (e.g. Serum) as the synthesis signal. Is this a correct, @tiagodb?

That cannot be done using EVOC, as far as I know. That is, again, because EVOC uses its own internal synthesiser only as the synthesis signal. 

But, as I said, please correct me if I am wrong or if it is I who have misunderstood something in the communication above. 

[oscwilde]

34 minutes ago, Dr.Socrates said:

That cannot be done using EVOC, as far as I know. That is, again, because EVOC uses its own internal synthesiser only as the synthesis signal. 

True, but the sounds in the video can easily be generated with the EVOC20 PS.
Other tools in Logic can also do this, but using automatic pitch tracking - rather than keyboard input to define the pitch of the synth signal.
The EVOC 20 PS actully has 2 synth engines....FM and Dual mode - the latter can use any of 50 single cycle waveforms for each oscillator.
In FM mode, typical carrier / modulator pairing of sine waves.
There's also a noise source that is available for "unvoiced" signals (fricatives, a.s.o.).

The way that Ableton approaches things - basically separates the Analysis engine into it's own plug-in....and allows MIDI note input from a clip to control pitch. The output is sent to whatever synth engine you want to use.
While there is flexibility to this, I don't believe the Ableton vocoder can accept live MIDI input (perhaps someone can clarify?)

The Live vocoder is probably most equivalent to the EVOC20 TO plug-in....although the Logic plug offers internal scales/individual notes can be removed/added/tracked.

Bottom line...talking/singing robots, Darth Vader/Mickey mouse, or trashing drum loops, etc....definitely possible/easy to do in both apps.

[Dr.Socrates]

Thanks, oscwilde! 

My bad, thinking only of the EVOC PS! You are right, oscwilde, that using the EVOC20 TO (an audio fx plugin) it should be possible to do what tiagodb was asking for! Thank you a lot for correcting me, @oscwilde, and sorry to you, @des99, that I was misunderstanding your response (which I now realise that I was – thinking again about what you meant with "it comes in three versions")!

@tiagodb, here is an easy example of how to do what I think you were asking about: 

1. Open the synth of your choice on (e.g. Track 1). This will be your synthesis signal. Load EVOC20 TO as an audio plugin on this track. 

2. On e.g. Track 2, place the audio material you wish to use as analysis signal. (You could also use live material, in which case you should choose the corresponding input as the side-chain in EVOC 20 TO, see below).

3. In EVOC20 TO, choose Track 2 as your side chain. Then, use "Side Chain" as analysis in, and "Track" as synthesis in (see picture below). 

Press play. 🙂

(Again, sorry for my mistake!)

 

Edited March 10 by Dr.Socrates

[tiagodb]

18 hours ago, Dr.Socrates said:

I am quite possibly wrong here, because I know that des99 is very often right, but I think that there is some misunderstanding going on here, and that it is in fact not possible to do what tiagodb asks for using the EVOC. Please correct me if I am wrong!

First, some terminology, because it is a bit difficult to understand what you are asking for, tiagodb. As per the Logic manual, a vocoder basically has two main ingredients, an “analysis signal” and a “synthesis signal”. In EVOC, you can e.g. use an audio input, a pre-recorded audio, the output of some other track or a bus as the analysis signal, which is routed in to EVOC using the side-chain setting, as des explained already. So, des is correct that you can use your own voice, or the output of some other synth (Logic's own or 3rd party) as the analysis signal. Or a drum beat. Or whatever sound you like. But, if I am not mistaken, EVOC always uses its own internal synth for the synthesis signal. 

Now, I think that what tiagodb asks for is to be able to use one audio source (e.g. a voice) as the analysis signal, but another external sound source (e.g. Serum) as the synthesis signal. Is this a correct, @tiagodb?

That cannot be done using EVOC, as far as I know. That is, again, because EVOC uses its own internal synthesiser only as the synthesis signal. 

But, as I said, please correct me if I am wrong or if it is I who have misunderstood something in the communication above. 

 

Thank you for all the suggestions and attention to this. Regarding your question, that's exactly it. I want to use presets from Serum, for example, to modulate the Vocoder.

As for the @oscwilde, indeed the EVOC can generate sounds like in the video, but considering the plethora of presets from Serum/Diva/any other synth, my possibilities for creating a unique Vocoder sound increase dramatically.

Thank you for your attention. I will try out new steps.

 

[Dr.Socrates]

Just now, tiagodb said:

Regarding your question, that's exactly it. I want to use presets from Serum, for example, to modulate the Vocoder.

OK, great! Then try out the steps 1–3 that I suggested above, because I think that will let you accomplish precisely what you want. 🙂