Fatten cheap mics

[Andrew Johnstone]

Hello Chums,

I am wokring on dubing & relaying VO tracks a series of films for Africa for a development agency.

Many of the VO rcordings have seeminlgy been recorded on cheaper mics. I have now done dozens of thses short films and I have experiemented with several ways to fatten the sound which often seems to lack presence and tone. 

There are some very handy presents in the various standard Audition EQ tools, I am using some compression and other filters for pops, clicks and noise etc. i am using a rboadcast compressor on the master bus and I can manage to have things sounding 'OK".

But, in the never ending quest for decent tone, I'd really like to hear any tips & tricks for sewateing up cheap mics. I appreciate I can't replace what is not there, but there maybe a few nuggets of info out there...

Thanks

[EuanWilliamson]

Hi Andrew,

Can you post us a small sample, max 30 secs, of what you have to contend with so we can take a listen 🙂

[Andrew Johnstone]

Thanks for the reply. Here's a clip - well a few clips bundled together form a scene where a couple are chatting/arguing. There are no FX on this clip, no compression & no EQ.  The film sis about contraception & family planning.

You can tell from the performances that there are clearly variations in the actor's position on the mic (eg the shouty bits) there's lots of room/reverb and I am not sure what the singla chain was, but I should imagine that there was pretty much no compression or EQ running before it hits the tape - as it were.

I have had dozens of these short films to work on and each one presents a differnt challenege. I have a retty good process running and as I said depending on the qulity of the input, I can usually get the thing sounding OK, but I realise there will be better dubbing mixers than me out there and I'm always keen to learn...

[EuanWilliamson]

Thanks Andrew,

I'll take a listen and get back to you

[EuanWilliamson]

Hi Andrew,

I'm not sure of course what you've tried.

I had a look and listen to the file an used the frequency analysis on a section.

It shows a general lack of presence in the bass frequencies.

This analysis was of the section highlighted in the next pic.

 

A correction made with the FFT filter ( or you could use the parametic equaliser ) boosts the bass frequencies

to counteract the lack in the recording. Perhaps combined with your other "fixes" this may help.

Of course as you say, you have many variations of many mics so it's not possible to find one solution that fits all 😞

[Andrew Johnstone]

Thanks for taking the time to have a look Euan, I appreciate it.

I have done a lot of bog standard audio dubbing for broadcast TV/docs and as a rule, for a lot of factual content, you don't need to do much more that ensure that the content is audible (!) and the basic tools in FCPX/Premier or whatever are more than adequate to sweeten things up and meet TX requirements.

This project as been a bit more advanced. Some of the files that have been sent in have been bang on. Lots of presence in the recordings, a broad tonal/frequency range, little room tone, so very easy to work with. Other files have been frankly shocking, but the cavet is that that some radio stations in rural parts of africa are no kitted out like a pro studio in London or whereever.

I have been able to fix most things, but those recordings where you can clearly hear that the mic quality is poor are very tough to work on. One trick I have used on a few tracks has been to double them and EQ each track differently (one gets a lot of bass and the other a load of treble for example) and this works quite well. I gues I am just looking for that magic bullet!

[SuiteSpot]

It is very difficult if not impossible to add something that was never recorded in the first place

[Andrew Johnstone]

Indeed. But the really top knob twiddlers have their own brand of magic sauce...I am just trying to lure them out of their hidey holes for some top tips...!

[SteveG_AudioMasters_]

My top tip is to try a Mic Modeler... Best microphone modeling plugins 

SuiteSpot's right, though. And it's not just about what wasn't recorded, it's also about distortion. And it is absolutely impossible to fix distortion, because you have no reference to what it should have sounded like - and if you had, you wouldn't need to fix it anyway!

[Andrew Johnstone]

Mic modelling, now that's a top tip.

[EuanWilliamson]

I hadn't heard of mic modeling before - very interesting.

I'd heard of amp modeling and I guess I should have realised that other types of emulation would exist !

Would I be right in assuming that mic modeling will have issues with sub optimal input?

What's your experience of these please 🙂

[SteveG_AudioMasters_]

Would I be right in assuming that mic modeling will have issues with sub optimal input?

By @EuanWilliamson

In a word - yes. The original Antares mic modeler (about the only one you could get without an actual microphone to start you off) came with a warning that it wouldn't turn a SM57 into a U87. Basically it's a convolution system that they use - so you obtain an impulse response from your source mic and by convolving it (with a few twists) with a different mic, you end up with the sound from your original mic coming out a bit more like the sound from the one you chose.

Any distortion is going to remain - and if you use a cheap dynamic and try to emulate a decent condensor, then clearly you aren't going to get any of the characteristics that the real one would give you. And on top of that, because of the nature of the way you have to capture the impulse, the whole system is rather position-sensitive. So no moving closer and getting a bass boost, no sense of 'air' around the sound, nothing like that.

Back in the days of the original Syntrillium forum (the one that became AudioMasters) we had a huge row with the guy trying to create these emulations, because at the time he insisted that you could mimic a U87 with some ropey old Shure dynamic. He pooh-poohed everything we told him that said you can't do it (all based on common sense) until he actually tried to do it. That wiped the grin off his face somewhat... 😉

So no, it won't produce wonderful results from a crappy input. But sometimes it can create a sound that's more acceptable (just...), so it might be worth trying anyway.

[EuanWilliamson]

Thanks @SteveG_AudioMasters_ 

[chrisw44157881]

EQ first, maybe raise 150hz-200hz. and play with 1150hz, and 4-5khz.

also, sounds like you have some distortion in there.

to get a fuller sound, use multiband compression  and enhance certain freq. don't go above 2:1 compression.

mic modeling is to try to make one mic sound like another. you can record pink noise in both mics and match them up in freq graph if you want, but ear is where you want to go.

some high end tools just automate multiband compression but they aren't really 'better'.

there's also transient modelers, but I didn't hear too much high end. just wanted to let you know they exist.

also, for really bad mics, you can add micro studio reverb at 150-170hz which is what some mics do to give warmth. has to be used sparingly and only for isolated ranges.

[Andrew Johnstone]

That's all really helpful. The batch of audio I am wrest;ing with this week is even worse, but all these tips are handy.