I'm experimenting with putting my vocal through a vocoder, but I don't really get how to do it.
All I can so-far manage, is to vocode the vocal on one note only, whereas what I want is to vocode every note individually, so it matches up to what is being played.
Does that make sense? - I want to maintain the melody and the tuning of the notes, but just vocode them to create an interesting effect. - How do I do this?
I have a feeling that the answer may lie with "sidechaining", something which I also don't fully understand!
I'm using Logic, and I have access to either the AKAI DCVocoder, Steinberg Vocoder, or the free CM Vocoder that comes with the Computer Music magazine.
If someone could show me how to achieve my desired effect using this software, I would be very grateful!
Vocoding - don't get it...?
-
JumpingJackFlash JumpingJackFlash https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=44005
- KVRian
- 1227 posts since 10 Oct, 2004
-
- Skunk Mod
- 21249 posts since 10 Jun, 2004 from Pony Pasture
Feed a tune into the carrier input and your voice into the modulator input. Voila -- the singing computer sings your tune. (DC Vocoder uses left and right channels for the carrier and modulator [or vice versa, I forget] so you don't need a sidechain to get that second input. A stereo vocoder would have to have a sidechain in order to get both signals in there.)
[edit] Depending on which vocoder you use, you might have to switch off an internal "drone" carrier source in order to feed in your melody.
[edit] Depending on which vocoder you use, you might have to switch off an internal "drone" carrier source in order to feed in your melody.
-
- KVRian
- 520 posts since 13 Aug, 2002 from Salzburg, Austria
You cannot do this, you need a sequence on the carrier, if not you'll end with one single note.I want to maintain the melody and the tuning of the notes, but just vocode them to create an interesting effect. - How do I do this?
If you plan to to an exact melody, I suggest you to transpose the vocals a second or third down using a pitch shifter because if both modulator and carrier are playing the exact notes, there will be a harsh resonance.
-
- KVRAF
- 8706 posts since 24 May, 2002 from Tutukaka, New Zealand
And here's an actual how-to-do-it quickie.
(I use Cubase on a PC, so it may be a little different, but the idea should be the same in Logic I'd hope...)
On your vocal channel you insert a vocoder. The vocals feed the modulator input.
Depending on your vocoder - some have an in-built sound source for the carrier (which makes it easier for starting up with) you route the output of a midi track to the vocoder. This feeds the carrier signal in the vocoder - i.e. the bit that actually plays the tune on the vocoder. In Cubase, on every midi channel you can route the midi to whatever output you want - i.e. you can feed the midi sequence to any active VSTi or to any active FX that can receive midi (which in this case is the vocoder). For a vocoder, you simply route the midi to the vocoder.
So you basically need 2 channels - 1 - your vocals; 2 - a midi sequence to play the vocoder.
Does that help any more?
With a vocoder that can have any other audio source routed to it, what you do is switch the vocoder to audio input and feed the output of another audio channel or VSTi into the carrier input. So, say you want a PWM/white noise chord sequence to play your vocals, you still have the vocoder inserted onto the vocal channel, and route the output from the chord channel into the carrier input of the vocoder.
You have to get your chord pattern, or your midi sequence to sync with where your words in the vocal recording are. If your note cuts off too early, then your vocoded vocal will cut out early too. If your note is too long and runs into the next word, then your vocoded vocals will also change pitch half way through the next word.
As I say, I use Cubase. I always found Logic to be a weird host that does things arse-upwards, so it may be different in your case. As mentioned, it's possible you might need a sidechain to get stereo vocoding from an audio or VSTi source. But certainly something like MDA talkbox (and there are a couple of other vocoder plugins that are based around the Talkbox) has an inbuilt carrier wave, so try that way out 1st before you get into sidechains. Any vocoder that has an inbuilt sound source will tend to use tri waves, or saw waves mixed in with noise, so that there is plenty of frequency in there to play with. (With a vocoder, you tend to get better results with sound sources that have a full sound - things like sine waves don't generally work that well, although they can depending on what you want).
Similarly if you want to use a drum loop to make a gated pattern on a chord sequence - you insert the vocoder onto the drumloop channel and play your notes in the same way as you would with the vocal example. Then the chord sequence will play in the rhythm of the loop but with a vocoded sound.
If Logic does it a different way - maybe Sascha here can help (He's got a degree in alternate reality...i.e. he's a Logic user
).
(I use Cubase on a PC, so it may be a little different, but the idea should be the same in Logic I'd hope...)
On your vocal channel you insert a vocoder. The vocals feed the modulator input.
Depending on your vocoder - some have an in-built sound source for the carrier (which makes it easier for starting up with) you route the output of a midi track to the vocoder. This feeds the carrier signal in the vocoder - i.e. the bit that actually plays the tune on the vocoder. In Cubase, on every midi channel you can route the midi to whatever output you want - i.e. you can feed the midi sequence to any active VSTi or to any active FX that can receive midi (which in this case is the vocoder). For a vocoder, you simply route the midi to the vocoder.
So you basically need 2 channels - 1 - your vocals; 2 - a midi sequence to play the vocoder.
Does that help any more?
With a vocoder that can have any other audio source routed to it, what you do is switch the vocoder to audio input and feed the output of another audio channel or VSTi into the carrier input. So, say you want a PWM/white noise chord sequence to play your vocals, you still have the vocoder inserted onto the vocal channel, and route the output from the chord channel into the carrier input of the vocoder.
You have to get your chord pattern, or your midi sequence to sync with where your words in the vocal recording are. If your note cuts off too early, then your vocoded vocal will cut out early too. If your note is too long and runs into the next word, then your vocoded vocals will also change pitch half way through the next word.
As I say, I use Cubase. I always found Logic to be a weird host that does things arse-upwards, so it may be different in your case. As mentioned, it's possible you might need a sidechain to get stereo vocoding from an audio or VSTi source. But certainly something like MDA talkbox (and there are a couple of other vocoder plugins that are based around the Talkbox) has an inbuilt carrier wave, so try that way out 1st before you get into sidechains. Any vocoder that has an inbuilt sound source will tend to use tri waves, or saw waves mixed in with noise, so that there is plenty of frequency in there to play with. (With a vocoder, you tend to get better results with sound sources that have a full sound - things like sine waves don't generally work that well, although they can depending on what you want).
Similarly if you want to use a drum loop to make a gated pattern on a chord sequence - you insert the vocoder onto the drumloop channel and play your notes in the same way as you would with the vocal example. Then the chord sequence will play in the rhythm of the loop but with a vocoded sound.
If Logic does it a different way - maybe Sascha here can help (He's got a degree in alternate reality...i.e. he's a Logic user
).