Do you use a compressor, EQ,

D

deleuze3

Well-known member
Hi,
1) for narration, do you use a compressor, EQ modification, some reverb In Adobe Audition ?

2) I use a multiband compressor with the margin at -3db. The narration is at good level, but it seems to add noise and sound a bit too compressed.

I tried to adapt to -16 LUFS, but it gives me a couple of high peak with distortion.

Any tips ?

thanks in advance,
deleuze3
 
For dialog, I would normally use.. a 5 or 6 band parametric or paragraphic EQ w/ hi and lo pass filters, a single broad band or multiband compressor and a brick-wall peak limiter... (there are no one size fits all compressor presets) and whatever else is needed. Too much of anything will usually destroy a track. I never use reverb... aside for special effects or to match ADR.
IMO, track volume envelopes are our best friend.
 
Last edited:
Once again, you are looking for a "one size fits all" solution to your sound, much as you did in THIS THREAD a couple months ago.

And again, there is no one-size-fits-all formula or preset for all of your sound!

Have you tried researching what a compressor is, what it does, and how each setting affects the signal? Understanding those things will help you to determine when and if a compressor is needed to achieve the desired outcome.

Compression reduces the overall dynamic range of a signal; that means it reduces the distance in dB between the softest sound and the loudest. To do this, compressors generally have a few standard settings:

Threshold is the dB level at which, and above which, the compressor affects the signal. If the signal is below the threshold, the compressor is inactive. The threahold setting will differ for every source, based on how much of the signal you want to compress.

Ratio is how much the compressor affects signal at or above the threshold. The ratio is represented by dB in to dB out. For example, a ratio of 3:1 means that an increase of 3db above the threshold on input will be output as an increase of only 1dB. The higher the ratio, the more the signal is reduced (compressed) before output. The ratio setting will differ for every source based on how heavily you want the affected signal to be compressed.

Gain is also known as makeup gain. This setting affects the overall output after compression, so it increases the lower, uncompressed signal as well. The gain setting will differ for every source based on how much compression was used, and how much more you need to bring up the quietest parts of the signal.

The other two standard settings are attack and release. Rated in miliseconds (mS), these determine how quickly the compressor kicks in after crossing the threshold (attack) and how quickly the compressor opens back up after squeezing down (release). These settings will also differ depending on each source and how it is compressed. And if you see a knee setting that selects between hard and soft, that determines how sudden the compression occurs. Hard knee clamps down instantaneously, while soft knee has a sloped response. Hard knee is not something you want to use in dialog.

By compressing the peaks above a certain point, and the using makeup gain to increase the overall level, the peaks are squashed and the dips are raised up, which minimizes the difference between highest and lowest signal levels. Compression can be used simply to tame sudden peaks, or it can be used to make a source sound louder (perceived loudness) for a very bold, in-your-face kind of sound. The heavier the compression, though, the more it affects the signal's frequency response, specifically in the higher frequencies. This is why we put EQ after compression, to be able to correct for any artifacts caused by compression.

Is it a good idea to use a compressor? Maybe, maybe not. It relly depends on the source material and where/how it needs to sit in the overall mix. Same goes for EQ. And if there's noise in the recording, compression will likely exacerbate the issue (in other words, the compressor will make the noise much more audible).
 
Last edited:
deleuze3,
If your looking for a more up-front 'in-your-face' type sound you're probably SOL. Your clip 'sounds' like it was recorded with the microphone too far away from the source (the person speaking) The recording space/location and gear can also play heavily into this that no amount of processing can compensate for.. (neither can expensive top-shelf mic/preamp/recorders)
Alex is right on describing compressor parameters and it's usage/results. The processor chain order is subjective though.. Again: 'one-size-does-not-fit-all'
 
no i don't want the sound to be more "up front".

For the video, i used a compressor. Maybe i shouldnt in this particular case.

Thanks Rick and Alex for your advices
 
deleuze3 said:
Hi,
1) for narration, do you use a compressor, EQ modification, some reverb In Adobe Audition ?

2) I use a multiband compressor with the margin at -3db. The narration is at good level, but it seems to add noise and sound a bit too compressed.

I tried to adapt to -16 LUFS, but it gives me a couple of high peak with distortion.

Any tips ?

thanks in advance,
deleuze3
Click to expand...

I know people hate it when folks reply.... It depends.

But really, it depends.

I'm not up on Audition, so I won't address answers specific to it's toolset, but most nonlinear editing systems should offer about the same basic tools.

The majority of my narration work goes out clean, no compression, no EQ, and completely dry with no reverb. Just my voice into Mic/Pre, with no additional plugins applied. For example.. a U87 into a John Hardy M1.

Then I have clients who want a more produced promo voice track delivered to them. So I might patch in a compressor to track with. For example.. an MKH416 into an Avalon Vt737sp into an Empirical Labs EL8 Disstressor.

It's only when I'm delivering a fully produced piece that I will add compression, EQ, Limiting, and the like inside of Pro Tools.

I rarely add EQ while tracking, as it's not needed tracking in my booth.

The exception, is when I'm traveling. I'll sometimes need to knock out a VO while on the road, and hotel rooms and various locations can be a *****. So I'm adding a little 500 series rack that Aphex is blowing out at $299 right now for my new portable rig. While 4 slots is not many in studio use, it makes it perfect for traveling. A Mic Pre in slot 1, De Esser in slot 2, EQ in slot 3, compressor in slot 4, and it should be helpful for taming some of these bad locations.

Dave
 
deleuze3 said:
https://www.youtube.com/watch?v=4T-sfqzMCXk&feature=youtu.be
Click to expand...

Rick R said:
Your clip 'sounds' like it was recorded with the microphone too far away
Click to expand...

To me, you don't sound too far away. I hear some echo, but it sounds to me more like hard surfaces near you. This is why pros record in sound booths. As a cheap shortcut, I've heard of news reporters recording their narration in bed under a blanket.

But to me your sound is not bad, especially compared to most stuff out there recorded without a dedicated sound professional.
 
I watched the posted clip - I hear that compressor - breath sounds that should fade down are being pushed up -they fall into the compressor threshold and become more prominent - the high is a little harsh vs. silky. There's no low end to speak of, it's all mids - VO should be "the voice of god", the voice in your head... usually more bass differentiates this from dialogue/naturalistic sound. Having someone who can speak close to a mic without popping it (esp. a large di. mic) sounds "closer than close" - again, "inside your head". I hear a bit of the room in there too - I feel like I'm hearing some hard walls.

I'm by no means a pro sound guy - so here's my "non-pro" take on it...

Your #1 battle is capturing the sound. Mic close (but not too close or pops and breath will kill you). Gain properly staged (even mic placement is to some extent "gain staging"). Getting the most level you can with safety overhead for peaks. Low cut at the mic (preferably) or the recorder if needed. An unobtrusive limiter if things are a little too run & gun to monitor & re-do for overs and pops.

Environment - QUIET and non-reverberant. I've heard VO's cut in (parked) cars that sounded good in a pinch. Moving blankets can make a big difference, but surround the talent. If you're cutting VO on location, this can all be really tough. If you're doing an on-camera interview, it can be even worse. RECORD ENOUGH "ROOM TONE" to have something for noise reduction to sample!!! Get someone to be the "quiet police" if you're at a corporate or industrial gig - the client will usually know just the person.

Even a $100 Sure 57 will give you good audio if you start with the above. Move up to a decent hyper (on set especially) and you should be able to reduce any ambiance. Get a really good hyper and you'll capture more breath & crispness. If you can treat a room for voiceover, there's nothing like a large diaphragm condenser, esp. for male voices. (Those wrap-around mic deadeners you stick on a stand can really work wonders, esp. if you can have a padded wall behind you). If you're on the cheaper side, a Sure 58 or other dynamic can give you a lot of bass control via proximity. Closer = more depth.

From there, it's what are you hearing vs. what do you want to hear? Noise/hiss/BG sounds... even free Audacity can help. Noise reduction (which can really mess up a track if overused), de-essing and gating (same risk of trashing a track) can clean things up. Manually adjusting levels (again, it can be overdone). Get the tracks as clean as you can. Sometimes, you just have to have simple music to mask location sound in post. It sure can help when done right.

From there, if you want a bolder, more in-your-face sound, compression can help... having a choice of compressors is nice - some of the vintage-emulated stuff can add some really nice mojo and sparkle. so can EQ (and again, a modeled EQ can do things close to what an exciter seems to do). But you have to LISTEN to the efx, cut them in and out, etc. and see if it's what you want.

I've even pitch-shifted tracks down a hair for more "gravitas" (and when I write the script and record a scratch VO of myself, I do this all the time - about 30% of the time the client says "no, that's great, we don't need to buy a voiceover").

Editing is huge as well... rebuilding a fluffed word, killing stutters, and time-stretching things can make a big difference. It's all a chain that can be trashed by its weakest link.
 
Thanks everyone for your advices. I highly appreciated

I first did test with a Oktava mk012 directly in my camcorder Sony HXR-NX30 but the sound is thin and is not fuller enough.

I finally record the video with Sony HXR-NX30. I record the audio with the OKTAVA mk012 plugged into a Sony pcm-m10, which is plugged into a Sound Devices MP-1.
(the level of the pcm-m10 is at -12)

In post-edition for audio, i only raise the volume. No compression, no EQ.

I think it sounds good.

Thanks for your advices, it gave me ideas.

Here is the results :

https://www.youtube.com/watch?v=hT_MOcNn3xA&feature=youtu.be
 
That sounds decent on my laptop. It would have been a little better if the mic was closer, but due to the (strange to me) camera angle, that probably wasn't possible.. I did hear some extraneous background noise through. (music and such) I'll check it again when I get back to home base w/studio monitors.

Addendum: Sounds OK on my monitors as well. The extraneous background noise (music's low frequencies beat) are audible, but low and should not be a problem. If the mic was closer to the source this would reduce the BG noise a little and give the voice more punch as well.
 
Last edited:
For what you are doing it sounds fine. If it were a VO for a film say, it doesn't have that CU "sound of god"" quality and there is too much room on it. But for a video leter it sounds like it's in the room and it's not too bouncy and the quality of the sound seems good.

As a side note an up shot like that tends to give it a slightly "sinister" look which may not be what you want.
 
Rick : thanks for the comment. Thanks for possibly check it again when you get back to home base. If there is other things you noticed or tips, let me know.

Noiz2 : Thanks for the comments.
How do we make a CU "sound of god" quality ? (You mean when Morgan freeman do voice-over ? A great actor, and maybe by using a condenser studio mic and a bit of effects?
How can you eliminate a bit of room in audio ?

Yeah the camera angle is a bit strange, i agree.

Have a good day, everyone that contributed.


P.S. Just to compare i did another version where i had the Oktava mk012 plugged directly into the camcorder and the sound was not as good.
It sounded thin, less amplitude, less bass, it sounds that its a bit distorted. What do you think ?
https://www.youtube.com/watch?v=4nlPavRDumA

Compare when i plug the Oktava into the Sound devices (the original video for this forum discussion that you already heard):happy: :
https://www.youtube.com/watch?v=hT_MOcNn3xA

 
Last edited:
deleuze3 said:
How do we make a CU "sound of god" quality ? (You mean when Morgan freeman do voice-over ? A great actor, and maybe by using a condenser studio mic and a bit of effects?
How can you eliminate a bit of room in audio ?
Click to expand...

For one, you've been a bit all over the map on this... again. The very first video you linked was VO (voice over video, or narration), and did not seem to be intended as on-camera dialog. Yet, it sounds like it was recorded with the mic placed for on-cam dialog. Your follow-up videos were shot as on-cam dialog.

It's really tough to steer you in the right direction when you leave no continuity every time you post another attempt. So let's clear up some things:

If you are recording for on-camera dialog (that means that the person speaking will be seen on-screen as s/he speaks) then mic placement is completely different that recording VO/narration (which means the voice is the only thing recorded, intended to be used over other video footage).

If you really are trying to record VO or narration, then you need to be right up on the mic (within 6"), and in a very quiet, non-reflective room. That proximity, being very close to the mic, is what helps minimize extraneous sounds since the direct sound (voice) is closer and thus at a much higher level relative to anything in the background. THAT is how you get the close-up, "voice of God" quality.

If you're going to sit in front of the camera, and record while speaking to the camera, that is NOT VO.

And again, compression is used only when absolutely needed. Same with EQ. And reverb? Only if that's the effect you want to hear.
 
You must log in or register to reply here.
Back
Top