Which vocal effect would you want ?

[ponce]

Hi people of KVR,

I'm tinkering with vocal processing and figured it would be a nice input to ask here for opinions.

Are you satisfied with existing vocal effect?
What would you want to do with vocals which you currently can't?

Feel free to ask for anything

[schnapsglas]

You know, everything is fine, except the dialogues in post situations. But if I wanted a plugin, I would want something that tracks the pitch and boosts certain harmonics accordingly. This would be difficult as it depends too much on the time domain, but is better than doing one word at a time.

[ponce]

@schnapsglas; thanks, I'm taking notes while hoping more serious plugin developers don't notice.

[theodore_whitmore]

For me the 'killer app' would be something that takes out the ugly, hard room resonances that seem to plague vocals recorded in less than ideal circumstances. Conventional EQ can harm as much as it helps and dynamic EQ is a big old PITA to work with.

My second choice would be a reverb that adapts to fit a vocal. Good singers adapt their singing to best excite the natural acoustics of a room. Even good digital reverbs can sound tacked on to a vocal. A classic modulated hall/room reverb that follows the character of a vocal to create the sense of a well stimulated acoustic space would be useful to me.

[schnapsglas]

What about a mic-to-singer proximity sensor? Would be useful...

edit: that gives me an idea. One or two parameter plugin that changes the perceived proximity with automation support?

[sws2h]

What about a mic-to-singer proximity sensor? Would be useful...

edit: that gives me an idea. One or two parameter plugin that changes the perceived proximity with automation support?

Little Labs VOG goes this (or sounds like it). UAD only though.

I'd like a good Automatic Double Tracking (ADT) effect.

[ponce]

What about a mic-to-singer proximity sensor? Would be useful...

You mean to avoid the proximity effect?
Looks like a dynamic bass EQ would do the trick.

edit: that gives me an idea. One or two parameter plugin that changes the perceived proximity with automation support?

Not specific to vocal, but there is http://www.tokyodawn.net/proximity/ already (which I think works fine coupled with a binaural pan).

[ponce]

For me the 'killer app' would be something that takes out the ugly, hard room resonances that seem to plague vocals recorded in less than ideal circumstances. Conventional EQ can harm as much as it helps and dynamic EQ is a big old PITA to work with.

Interesting use case, but to be honest this looks a bit hard for me and not very specific to vocals. I think there is a blind deconvolution plugin which might be able to remove the verb. I can't find it right now.
Imho this problem is with the room reverb and not specifically the resonances.

EDIT2: I'm not sure there is anything special in a vocal sound to make reverb removal easier.

My second choice would be a reverb that adapts to fit a vocal. Good singers adapt their singing to best excite the natural acoustics of a room. Even good digital reverbs can sound tacked on to a vocal. A classic modulated hall/room reverb that follows the character of a vocal to create the sense of a well stimulated acoustic space would be useful to me.

Interesting. I can see how one could remove annoying reverb over-activity when doing "ssssss".

EDIT: can you be more specific about what good singers would do to excite the natural acoustics of the room?

[Tricky-Loops]

A revolutionary vowel or/and formant filter would be great. There are some filters out there, but they all sound similar. I think of an effect that can do more than the usual "e", "a", "o", "i" and "u" processings... Something like Angelina by Big Tick, but with much more features...

[ponce]

I'd like a good Automatic Double Tracking (ADT) effect.

I would have thought Antares Choir/Duo to be enough. Thanks.

[ponce]

A revolutionary vowel or/and formant filter would be great. There are some filters out there, but they all sound similar. I think of an effect that can do more than the usual "e", "a", "o", "i" and "u" processings... Something like Angelina by Big Tick, but with much more features...

Indeed many things to try there:
- more vowels by F1/F2 tuning instead of choosing a single vowel
- the "singing" formant when F3/F4/F5 are merged
- Choosing between male/female formants, the female ones are a bit higher in frequencies
- in real voiced sounds, formants which are near are more audible to help separation

[schnapsglas]

What about a mic-to-singer proximity sensor? Would be useful...

You mean to avoid the proximity effect?
Looks like a dynamic bass EQ would do the trick.

I mean some hardware device that tracks the movement during the recording session to ride the fader nicely and maybe even overdo this effect! But thinking little too away here.

[NAD]

Yeah, +1 on a good formant filter.
Something like Angelina as an effect would be awesome, though I'd like to hear a more pronounced (stronger) effect.
Also it would be really sweet to be able to sequence a string of arbitrary vowels and morph between them. For example instead of just I->E->A->O->U (Delay Lama style) you could also do O->I->U->A or whatever.

[ponce]

Yeah, +1 on a good formant filter.
Something like Angelina as an effect would be awesome, though I'd like to hear a more pronounced (stronger) effect.

FWIW I find Oli Larkin's Formant Filter more defined than Angelina's one. I like its LFO too.

Also it would be really sweet to be able to sequence a string of arbitrary vowels and morph between them. For example instead of just I->E->A->O->U (Delay Lama style) you could also do O->I->U->A or whatever.

IIRC since v2.0, Arturia MiniMoog has a formant filter effect which I believe you can use like a stand-alone effect. You can cycle through vowels in a pre-defined order and use voice-based automation curves to control the cycle speed (which can be negative). You are limited to that vowel order though.

[theodore_whitmore]

For me the 'killer app' would be something that takes out the ugly, hard room resonances that seem to plague vocals recorded in less than ideal circumstances. Conventional EQ can harm as much as it helps and dynamic EQ is a big old PITA to work with.

Interesting use case, but to be honest this looks a bit hard for me and not very specific to vocals. I think there is a blind deconvolution plugin which might be able to remove the verb. I can't find it right now.
Imho this problem is with the room reverb and not specifically the resonances.

EDIT2: I'm not sure there is anything special in a vocal sound to make reverb removal easier.

My second choice would be a reverb that adapts to fit a vocal. Good singers adapt their singing to best excite the natural acoustics of a room. Even good digital reverbs can sound tacked on to a vocal. A classic modulated hall/room reverb that follows the character of a vocal to create the sense of a well stimulated acoustic space would be useful to me.

Interesting. I can see how one could remove annoying reverb over-activity when doing "ssssss".

EDIT: can you be more specific about what good singers would do to excite the natural acoustics of the room?

Well, not being a good singer myself it's a little hard to say but you know it when you hear it! It seems they are able to force their voice slightly to stimulate the standing resonances in a room thus creating maximum acoustic reinforcement whereas lesser vocalists just put their voices out there to bounce round the room a bit. A good digital reverb on a good vocal can still sound like a voice bouncing round a room a bit. While putting a nicer, bigger space around the vocal it can make the performance seem smaller and weaker because it doesn't present a strong sense of acoustic reinforcement. It's definitely easier to hear than explain but I hope that makes sense.

As far as my first suggestion goes, I can understand that it's not an easy proposition to take on. However I'd guess >90% of recordings aren't made in ideally treated rooms and >99% suffer from room resonance issues. Taking out the most offensive resonances with some careful EQ makes a vocal much, much easier to mix. It sits in the mix more naturally so other sounds need less shaping to give it space. The natural expression and clarity of the vocal becomes unmasked so it needs less EQ to make it cut through. Compression 'fits' to the vocal more naturally.

To be clear, I don't believe it to be primarily a reverb problem. Some crude damping behind the vocalist will prevent the majority of reflections reaching the microphone but won't damp the resonances.

Secondly, you are quite right that there is nothing in vocals that makes it an easier case to treat. It is in fact a tougher problem. The ear is particularly sensitive to unnatural resonances in the human voice and to any measures taken to treat them. The fact that vocals are usually so upfront makes the case more pertinent.

OK, enough from me. I just noticed that you are GFM. I've found Psypan to be a useful tool. Many thanks for that and best luck with your future endeavours.