TAXI.com

OK guys, this has been on my mind for a very long time...probably won't get any replies but I'm gonna throw this out there... I've always been a massive fan of the French group AIR and their production sound-- but most specifically their vocal sound. It sounds very heavily processed in some way, but I have no clue just how. Unfortunately the rather generic name of their group "AIR" makes it impossible to do any web search that yields what I'm looking for.

Their album "Talkie Walkie" in particular has a vocal sound throughout that just blows my mind. The opening track "Venus" is maybe more straightforward-- but I wonder, is there some kind of heavy filter being applied to get that ethereal crisp sound? Is that just the timbre of their natural voices? Maybe a high pass filter?

But then the next track "Cherry Blossom Girl" has this ethereal crispness turned up to 11. What happens between a microphone and monitors to get that kind of sound? Or is it just their Frenchness?

Then the next track "Run". WTF? What is going on there?

Every song they do has an amazing vocal sound that doesn't sound human-- but it's not over-the-top weird-- I can't describe it. But I would like to know how to produce vocals in this style.

Here is the album to which I am referring: https://www.youtube.com/playlist?list=O ... 3smCoiIKfw

LIstened to the first few tracks.

Sorry, it's nice but I don't really hear anything majorly special/mysterious going on there ? If you want to sound like AIR then you have to write songs like they do and deliver them like they do as much as produce in a similar way.

It's probably just the result of a nice mic with a nice pre-amp for starters. Sounds pretty smooth...could be a ribbon ? The delivery is very held back, they aren't belting anything out it's more like speech level singing which means you can get a lot closer to the mic which sounds more intimate ( which would make sense if it is a ribbon mic as you can't be blasting loads of air at them ) . Also it's got subtle doubles in a lot of the parts I listened to so they don't sound like obvious doubles but they are mixed in enough to thicken the sound without sounding like a 2nd take.

Add a decent compressor, also sounds like there's some saturation going on and there's not really much in the way of reverb or delay..I reckon there's probably some sort of ambience patch rather than the more normal plates etc.

Part of the reason they stand out is because of the rest of the production is keeping out of the way, there's not actually that much going on when you listen to it. Sounds like an LCR mix so everything is either hard left/right or centre which gives more space to the central elements.

"Run " sounds like it's been sampled and triggered from a keyboard " run-run-run-run " ..that part is overlapping each other

Have you listened to any Billy Eilish ? I think there's similar stuff going on some of her vocal production and it's way easier to find info about her stuff.

https://www.youtube.com/watch?v=DyDfgMOUjCI

Mark

Mark, big thanks for listening to the examples and giving me your thoughts on what your ears tell you. It was that other thread on the Peer to Peer about the "Nu DIsco" vocal sound that made me think of what kinds of vocal sounds work for certain genres.

AIR does use a lot of deftly doubled parts as you noticed.

I'm still trying to find my voice as a singer and one of my frustrations is microphone technique-- setting the input level and knowing how close to get to the mic in conjunction with technique. Trying to make proximity effect work as a feature and not a bug. And also setting a compressor plugin.

I always wanted to be a baritone and have a breathy chest voice, but I took lessons for a while and alas I am a tenor and try to make the best of it. My best mic at the moment is an Audio Technica 4033a condenser. What drives me bonkers is sometimes I can get a really nice vocal sound when I'm singing a demo into my phone-- even if it's low fi-- but when I sit in front of the condenser mic, at my DAW I don't know why, but I struggle.

I've heard a lot of good things about ribbon mics for these types of vocals... this year hasn't been the best financially, so I don't know that a new mic is in the cards for a while. Anybody have any recs on a good ribbon mic that isn't too pricey?

When I'm working on my own material, my preference would be to sing in a quieter way, close up to the microphone. I've been playing around with some ideas for the listing based on the Johnny Cash cover of NIN "Hurt". I don't have a big deep resonant voice like Johnny Cash, so I'm looking for another type of sound that plays to the vocal tessitura I've been given.

I'll watch the link to BIllie Eilish's production (though I really find the whole aesthetic she represents to be... eh, not such a great thing-- I'll leave it at that. haha)

Thanks again!

Ted

Ted wrote: ↑

Tue Aug 04, 2020 11:04 pm

My best mic at the moment is an Audio Technica 4033a condenser. What drives me bonkers is sometimes I can get a really nice vocal sound when I'm singing a demo into my phone-- even if it's low fi-- but when I sit in front of the condenser mic, at my DAW I don't know why, but I struggle.

Nothing wrong with that mic, I have the 4040. What's probably happening is the phone isn't capturing everything , it will be pretty "filtered " in terms or hi and lo pass so it's all mid range-y ..it's probably really compressed as well and the bit rate will be lower than your DAW. So all you're kinda left with is the part that matters. There's nothing to say you can't use the phone recordings by importing them into the DAW either.

My guess its a decent mic and preamp. But Its a good vocal performance too.

Breathy, versus belting.

Compressed like hell. Multiple stages. Vocal is ridden for evenness after that / waves vocal rider. Some high end added but not over extended (like Natalie Imbruglia's "torn" is really extended high end). Naturally all the vocals are going to be high passed but likely not much above 100 hz on anything. Subtractive eq on the resonant peaks on each voice. It is pretty de-essed as well.

Pretty ambient to me. The drier instruments on the wings in Cherry make it less obvious.

The doubles are likely VocAligned and tuned as they are really exact doubles.

Cherry Blossom has an obvious female double 8va / harmony

Note that in a lot of the tunes there is a loud L+R competent that really drowns the vocals in the stereo mix (drowns in a good way) so they don't stick out. Listening in Mono is a little more telling.

Ted wrote: ↑

Tue Aug 04, 2020 11:04 pm

What drives me bonkers is sometimes I can get a really nice vocal sound when I'm singing a demo into my phone-- even if it's low fi-- but when I sit in front of the condenser mic, at my DAW I don't know why, but I struggle.

Any chance when you use the phone you are thinking "demo" and just don't care about it so much and just sing instead of putting it under the microscope and worrying about every little aspect?

So Mark I took your advice and watched several videos on Billie Eilish production techniques and got some great stuff out of it. There is one Youtuber who makes a lot of videos showing how to get the vocal sounds of certain artists (mostly hip hop) using nothing but stock Logic X plugins-- he had one on Billie Eilish so I went into Logic and step-by-step copied his recommended vocal plugin chain with the same settings on everything-- then I saved it as a user preset so I can always get it again.

Here's the video:https://www.youtube.com/watch?v=C_cQbxGbl0g

This is for her song Ocean Eyes. He also does a video showing how to get "Bad Guy" sounds but I haven't watched that one yet.

So I made a quick brief generic throwaway track in the style of AIR just to see if I could get that breathy vocal sound. It's just a practice exercise and the vocals were one-takes and they don't line up all that well-- it's a bit more difficult to line up vocal dipthong sounds such as a long "I"-- which is really an "Ahh" that morphs into an "Eeeee" sound. and there are some pitch issues and a better job could have been used on the compressors (especially on that upper harmony)-- I think I'm setting my ratio and threshold at a severe enough point to tame some of those transients but they keep poking up anyway. Maybe my attack was too slow.

Still, I'm taking baby steps here and I'm closer that I would have been yesterday.

So here's my first practice exercise: https://soundcloud.com/ted-w/airy-vocal ... Rd31kMOpuz

(yeah the vocals are way back in the mix too)

Andy, thanks a ton for also taking the time to analyze these sounds for me. As a noob producer, I am always in awe when guys like you and Mark share the things that must have taken you years to be able to discern. I'd never heard of Waves Vocal Rider but I watched a demo of it and it's only $50 and it looks like something that I could really use, I'm gonna pick it up.

I'm going to look into some of vocoder type things as well-- I haven't played with them yet, but I have NI's Mouth and Finger instruments and from what I gather, they can be used as vocoder-like effects on the voice. I might play around with some formant shift effects too-- that might get me lightness where I need it and weight when its called for.

I spent some time this morning watching some vids on various lower priced ribbon mics-- I was impressed with the Golden Age series and was thinking about going for one--possibly the active one or the tube version-- if for nothing else than if I got the passive version I'd likely fry it within a week of getting it by plugging it in without remembering to turn off the phantom power. I could just see myself doing that...

But then it also brings up the issue of a preamp and I used to have a cheap Presonus Tube Pre-- I still have it but it never sounded good to me and maybe it was a lemon or broken when I bought it. I was looking at the Golden Age preamps too, they looked nice for the money-- but as little as I know about mics, I know even less about preamps.

In response to the theory that perhaps my sound going into a phone is precipitated by a different level of mental comfort-- that could be part of it. I used to get serious fear anytime the record light would come on in my DAW. Not so much these days, though it is different listening to yourself through headphones while trying to perform.

I got a new interface a few months ago (NI Komplete Audio 6) and I am able to set my buffer much lower when tracking. My old ancient Lexicon had so much latency going on that I sounded like I was drunk when trying to record vocals-- and especially acoustic guitar or anything else with a heavy rhythm element.

Thanks again guys--

Ted

Yeah sounding better already...I notice the tutorial didn't mention vocal delivery which is a major factor..so maybe look that up next...

cosmicdolphin wrote: ↑

Wed Aug 05, 2020 2:30 pm

Yeah sounding better already...I notice the tutorial didn't mention vocal delivery which is a major factor..so maybe look that up next...

Will do. When I was taking lessons a few years ago-- and I really should have chimed in on that thread a few weeks ago about having the confidence to sing, etc (because what you wrote echoed some of what I went through) my teacher would constantly harangue me for the way I didn't want to open my mouth or move my jaw very much to sing. When I was supposed to sing "ahhh" it would always come out "uhhh" like the little upside down "e" used in the word "the"-- a "schwa" or whatever it's called.

I worked on my technique some more today.

Andy, I got that Vocal Rider plugin and it's freakin' fantastic!!! They were having a sale so I picked up a few other plugins too. I got a vocal suite and some freebies including a tape machine simulator that adds a really nice saturation to a mix-- and all sorts of other things. Also a pretty cool autopanner.

I just had a listed to Natalie Imbruglia's "Torn"... that is a LOT of high end. Never noticed that before.