I was listening to the B-side of “never ending story” 45rpm single by Limahl (Let’s face it, it’s basically just Giorgio Moroder with a young fella as a singer) which I think was called Ivory Tower, and I loved the vibe, but it was too positive for me, so I blended the feeling I was getting from listening to it when I was a kid, with a pretty depressing set of lyrics I wrote about facing an ever-shrinking future when compared to how it seemed when I first heard this in 1985.
I’m using the TAL Vocoder (VST) which is modeled on the sound of the early 80s vocoders (I was lucky enough to tinker with I think a Roland at one point 20 years ago that this sounds a lot like) but has a deceptively huge amount of range when you spend time with it. I’ve tried quite a few, but this one shines in the sound I’m looking for with that crunchiness that isn’t quite perfect. Being that I’m in Linux-land I use the linvst converter to allow me to use windows based VSTs.
I process it a little with EQ and compression to get it even (it can be wildly unpredictable) and tune the built in carrier generator with noise to get the breathy sound I want. I also sometimes use external synths to drive the carrier for more crunch.
I just grabbed the tal vocoder and I’m having trouble getting it to work in the mode where you use the built in sounds. I can’t quite figure out how to set it up. I have no problems when I want to use my own synth sounds and use a track group with and pan the voice and pan the synth. But for the life of me I can’t get the built in vocoder sounds to work. Could you describe your vocoder track setup for me? Thanks and sorry to hijack the thread!
Thanks - it’s a little sedate for what I like to make usually, but I still love it
So it’s a little tricky, but hopefully this can help:
Add the TAL vocoder to your auxilliary send (or add a new one if you’re already using the one created by default). It will be on the very right of your mixer (in mixing mode).
You’ll need two tracks. One will be for sample playback (or mic input), the other for the midi notes you’ll be sending to the vocoder.
In your mic input channel, add the “line input” object at the bottom (under the “routing” heading in the menu on the bottom left). Choose which input channel you’ll use for your voice in the object.
In your midi notes channel, this is where it’s a little tricky. Click anywhere on the midi channel you’re using, and then go to the right top corner and make sure you’re selecting an unused instrument (or insert a new instrument so it’s blank).
Now go to the “instrument properties” area directly below that and use the dropdown box. You should see a header for “VST FX alias” and under that when you press the arrow you should see your TAL vocoder object that you previously added to your send channel. This creates the ability to use the vocoder as an instrument to send notes to (MIDI note sending).
In your midi notes channel you can now (with the TAL instrument selected top right) record or just play notes that will send midi to the vocoder. If your mic was selected and on previously, you should now be starting to hear your voice being vocoded.
Some extra tips:
put a compressor on your mic input channel and tweak it so that your voice isn’t peaking much, keeping it fairly even. for me I keep the ratio about 1/3rd of the way from the left, and just bring the threshold down a little. You want your voice to be fairly even and flat, as vocoders become hard to understand if you have too much dynamics
put a compressor after your vocoder on your send channel. Again because of the nature of resonance with your voice pitch and synth pitch, you’ll get some mega peaks from time to time, you can play with what works without getting too robotic.
This is fairly easy for live playing/singing, but to record, you’ll want to go to the mic input channel, choose an empty instrument up top right, and then go to the sampler tab top middle to start recording your voice. Use the sync record to ensure you’re getting a synced sample to match your music. After recording, adjust the DC offset, and then use the “maximise volume of the sample” button to ensure you’re getting a nice strong signal from your sample to use. After this, you have to go back to your midi editor pane in the mic input channel, Turn off the line input, and then drop a middle C in the editor where you started recording with your sample instrument selected. On your computer keyboard this is the letter Q. This means the sample will play back as recorded (not faster or slower). You should now hear your sample playing through the vocoder (if you have midi notes also recorded in the vocoder midi channel sending to the vocoder)
In terms of settings on the vocoder, if you find you can’t get enough to understand words, add some of the noise channel (not too much, it’ll become a whisper), and I tend to just use the saw, and about half sub, with no pulse to get the clarity I need. I tend to run it in mono mode (it can get a little funky with too many hanging notes in poly mode) and I completely turn off the harmony. This means I get a good sound in the low end that’s a little grindy but still understandable.
In terms of sending midi to the vocoder, I tend to craft it so that there’s never any note offs, and I make sure I move all of the notes into the same column so that they are overwritten by each subsequent note keeping the sound clean. This means that the notes never stop, so whenever you sing, it’s sounding. With note offs you’ll get gaps which can be frustrating as we humans are less robotic than we aim!
I think that’s about it - hopefully that gets you onboard the good ship vocoder, and I’d love to hear what you get up to with it!
