Routine Dialog Post

[combatentropy]

Suppose you recorded someone talking. I'm mainly thinking about your typical Youtube video, someone talking at the camera from their living room. Suppose it had no problems. You had a good mic, good mic placement, no background noise, no weird hums, and the talent delivered the lines flawlessly.

What post-production steps would you always still do? Is there anything besides EQ and compression (dynamic-range compression)?

I've always lagged in my audio understanding. I want to give it the old college try. I still would be a one-man band, and I don't want to break the bank or break my back. You know with the right knowledge and a medium amount of effort you can get it really good. That's what I want. You could go further and spend hours and hours and hours to get it that last 1% of good. That's not what I'm interested in at this time.

I know there are probably interesting things you've done and clever rescues you've pulled off of some really messed-up audio, and a million plug-ins to choose from for every situation. Some day I would like to hear some of those stories. But right now I can't handle it.

I want to get an 30,000-foot view of the whole roadmap of audio post first, and I want to understand the simple scenario first (1 subject, simple dialog, no problems to patch) before I get into all the ifs, ands, or buts.

Here's my draft so far:

EQ

1. Notch out any room resonances (see my earlier thread, Surgical EQ, and thank you to all who helped me understand that part). Not sure if "surgical EQ" is the industry term for this step.
   
2. Roll off lowest frequencies, to taste. Could be conservative and just the bottom 100 Hz, could be like NPR and a gradual roll-off all the way back to 1,000 Hz (by setting the pad on the mic, I guess this doesn't count as post).

Compression

1. Ratio 2-3
2. Adjust other settings to taste (threshold, attack, release, etc.)
3. I could use some firmer guidance here, but from what I've seen it seems it could fill a small book all by itself

[combatentropy]

Also would be interested in how you categorize things.

I. Speech

A. Voice-over in a sound booth
B. Interview in a real space (living room, office, etc.) but perhaps treated for sound
C. Youtube direct address to the camera from home
D. Hollywood dialog

II. Music

A. Vocals
B. Instruments

III. Sound Effects


So you see I particularly am focused on just one of those right now (I. C.). But I may have mislabeled this thread Dialog, which makes you professionals first think I was going to talk about film dialog, which may have totally different post. I could imagine everything in the Speech section having roughly the same post steps --- or I can imagine them being widely different. I definitely understand the musical vocals are treated differently. For one thing, reverb is probably added to musical vocals but rarely to speech.

What words do you use for these different jobs? I don't think you use "Speech" for example.

[IronFilm]

I. Speech

B. Interview in a real space (living room, office, etc.) but perhaps treated for sound
C. Youtube direct address to the camera from home

I'd lump these two together. They're both an IV ("interview").

[Noiz2]

Also would be interested in how you categorize things.

I. Speech

A. Voice-over in a sound booth
B. Interview in a real space (living room, office, etc.) but perhaps treated for sound
C. Youtube direct address to the camera from home
D. Hollywood dialog

II. Music

A. Vocals
B. Instruments

III. Sound Effects


So you see I particularly am focused on just one of those right now (I. C.). But I may have mislabeled this thread Dialog, which makes you professionals first think I was going to talk about film dialog, which may have totally different post. I could imagine everything in the Speech section having roughly the same post steps --- or I can imagine them being widely different. I definitely understand the musical vocals are treated differently. For one thing, reverb is probably added to musical vocals but rarely to speech.

What words do you use for these different jobs? I don't think you use "Speech" for example.

Dialog, ADR, VO are the only "speech" labels I would use, though if I'm splitting things up I will generally put each character on their own track or two. If you are fixing word weirdness it's much faster to do it with two tracks than one.

Other than level balancing there is nothing I will always do to dialog. I will usually do a resonance sweep to see if there are frq.'s I should tame, but if it sounds good? Depends on the time line and the budget. Anything for the web producers always seem to want levels peaking at zero. I mix "correctly" and then do a final level and or compression on the output path.

IA - check for harmonics and then work with the director to get it as close to what they want as possible. VO in a doc. is fairly normal but VO say in a trailer is often hyped up.
IB - this is a situation where you probably are going to have harmonics you want to tame.
IC - that bar is very low as in absolute garbage seems to be acceptable to many. If I am doing it for me I want it as understandable and easy to listen to as possible. If it's for a client it kind of their call. In my experience they will probably want the levels crushed and compressed to hell and gone, but not always.
ID - now you are talking time. If it's low budget and I am the mixer I will usually not do a full checker board and fill air pass which you absolutely do need to do if it's going to a regular mix stage. in either you are going to be doing a LOT of slicing and dicing. Actually part of this also applies to all the above. In a hollywood dx edit you will separate all the characters to their own tracks and you will cut in fill air so that no track ever goes silent. The air track will hardly ever be used in the final mix it is mostly so you can play a track back for a director or producer and not have it sound choppy. They will never "unhear" that and make your life miserable or just fire you. In the actual mix the air will mostly be mixed out because the FX tracks are providing the "air" in any given shot. Which is why on a low budget gig I only add air where it is needed After I have the ambiences and BG's in and Never let a director hear a bare track. You will also go through the dialog removing moth clicks and various other unwanted sounds. You have to balance between too clean and too sloppy but taking out extra ums and such if you can with out weirding out the sync, ie you cant see lipps. Though some of this is more done on VO than dialog.
You said Hollywood, and I'm sure some things have changed but when I left dialog editors did very little with plugins. Things like VocAlign for ADR and that was about it. With, I assume, most shows now mixing "in the box" editors are probably allowed to do some basic non destructive things like maybe NR?
The reason is that the film is going to be mixed on a mix stage and the sonic choices really need to happen there not in a cramped edit room with questionable acoustics. Most dialog/ADR edit rooms have multiple editors working on headphones so not at all set up for making EQ choices.
But when I am doing "Hollywood style" editing on a low budget where I am mixing, which is probably closer to what you are asking, I have an EQ and a brick wall limiter on every track. I then bus all the dialog tracks to a Dlg sub master that also has an EQ and a brick wall limiter. ADR goes a similar route as does and VO.
The EQ's are on the tracks to correct recording issues and the limiters are there to keep me from having to worry about the unexpected explosive whatever. They are set at -3 and in the normal course of events are rarely active. They are "fuses" if you will, sound can pile up. It's kind of like level meters with some protection.
If luck is with me the EQ's will be mostly removing harmonic problems on the tracks and adjusting the over all sound on the sub. If there is a particular problem I may put a second EQ on the track to deal with it. For noise reduction I clone and mute the clip(s) and destructively use noise reduction. I have a bunch of other tools to "fix" things and what gets used where depends on the needs. There is often the need to make different takes of the same actor in the same scene sound the same. It's later in the day or maybe another day and they just don't sound like the same person so have to try and get them to sound like the same person.

I don't really do much music mixing. I do some recording for concerts and such but not really the mixing. Now for a film project I will do music editing, but that is really slicing and dicing.

SFX... There is a sticky that has a long piece I wrote and the thread that comes after covers that pretty well. My internet is a slug at the moment but I think it's called how to make your film sound like it had a budget. Something like that.

[Alex H.]

SFX... There is a sticky that has a long piece I wrote and the thread that comes after covers that pretty well. My internet is a slug at the moment but I think it's called how to make your film sound like it had a budget. Something like that.

It’s in the AUDIO 101 sticky.

[Noiz2]

Thanks Alex. I was going to put in a link but about then the internet here crashed. I didn't even think my post got up...
Now to try and get everything I need done before it all goes into the dark again!

[puredrifting]

Conversely, if you don't know what you are doing audio engineering-wise, be cautious and careful with EQ, compression, limiters and cut/pass filters.
Each of these tools, even at the rudimentary level that most video editing tools have, make it easy to mess up your audio in both subtle and dramatic ways.
Generally, a little goes a long way. And if you don't have a trained ear or a good ear, you may not be able to discern the levels you are changing things.
Just our eyes adjust and compensate white balance, our ears adjust and compensate for crappy sound unless you train them not to. It's easy to tweak an audio
mix, that to you, might sound "better" but objectively, sounds awful, until your ear has been trained to know what you are listening to and listening for. Location sound mixers
and especially sound engineers and studio mixers hear things in a different way than the average video person.

One other big factor, your audio reproduction capability. Do you have accurate, uncolored studio monitors in a sound treated edit bay? or crappy $20.00 ear buds? Or somewhere in between? Just as in doing CC and grading for picture, without a truly reference accurate sound reproduction system/environment,
you could easily be messing up your audio in ways that your system won't let you hear. Our editing system is hooked up to a decent mid level analog mixer running out to about $10k worth of Genelec monitors and a Genelec sub, the consolation prize when I shut down our audio post facility years ago.
The system has been tuned and equalized for the room we edit in. Many video editors are trying to do minor sound mixes using garbage, completely colored home stereo speakers, or commodity low end prosumer monitors. I'm not saying you cannot mix on cheap audio reproduction gear, I'm saying unless you have a starting point of "accurate/uncolored",
you are essentially flying blind (or in this case, deaf) unless you audience is listening with your same system, in a room that sounds identical to yours. And let's not get started about mixing on headphones, which is really only useful for playback on the same headphones but will typically sound terrible when
played back on speakers.

[combatentropy]

Other than level balancing there is nothing I will always do to dialog.

Thank you. It helps my understanding that you said that, especially knowing that you're a sound pro.

Other than levels, there was nothing I would do to dialog either. All my energy was spent just trying to get a good microphone, well placed. If I could do that, I didn't know what was left.

But toward the end of my career, on two separate occasions, two different people with audio backgrounds watched a video and said something like, "Hey if you ever want some help to make it sound really good, let me know." I thought the audio sounded fine, so I didn't know what they were talking about. After all, I got the microphones close. The second time, I took him up on it. I gave him the audio tracks, and a few days later he turned them back over to me, and I don't know if I could hear the difference. I don't know what he did. Probably brought down the low frequencies at least. Maybe carved out the music track's EQ to allow for the dialog. Maybe some mild compression.

[combatentropy]

It’s in the AUDIO 101 sticky.

Maybe I should read this

Feel free to keep commenting.

Or if you want to recommend books or articles, I'm game. I think I'll also peruse https://www.soundonsound.com/