The physics of sound and room acoustics – looking for answer

A

ARRIguy

Member
In an air conditioned room and without using a microphone (i.e. listening only with my ears) speech is typically acceptable and intelligible - even from someone on the other side of the room. Why, when I monitor sound coming through a mic, is the speech more smothered by the noise from the AC system and other ambient noise? Sure, different mics will record sound differently but there always seems to be some loss in clarity of speech compared to natural hearing (unless the mic is closer than my ears are).

Why should it be different and why should a mic emphasize the room noise so much over my natural hearing? This doesn’t just apply to viewing/listening to a video recorded image later (and viewing/hearing it out of context) but is the case even if I’m making this comparison in the room.

Can anyone someone give me a physics/acoustics explanation of this phenomena?
 
It has less to do with physics and more with psychoacoustics and neurocognitive functioning. Even the most expensive microphone is a crude facsimile of the human auditory system, not even considering the auditory processing that goes on in the brain. Human beings evolved the ability to pick out important sounds from any sort of background noise because it was/is imperative for survival.

All sound is the same to a microphone, there's no filtering or selection processing going on (other than the simplistic sensitivity pattern of a microphone). And in the process of recording, a huge amount of data is lost that helps humans localise sound in 3-dimensional space and process it.
 
Yeah, what the said.

In seriousness though; I've been on construction sites with the loudest equiptment and on a cutter in a storm and amazingly we could all communicate.
 
Exactly what Barry said.

Specifically the ear can tune out repetitive noise that it considers not a danger, yet can hear a very quiet sound in amongst all that dangerless noise that will instanly bring the ear's owner to full alert (e.g. a soft footfall in a garden with rustling leaves, for instance)

And as John says, the human voice goes from around 800Hz to 2500Hz and so if you filter out those frequencies in your non-dialog tracks, and/or very subtly (i.e. no more than +6db) boost those frequencies in the tracks that do contain dialog, then the dialog will be more audible - clearer.

I used to mix very loud disco music for a club in Manhattan and I was able to get a heart-stopping volume on the dance floor - and by virtue of filtering everything out between 1000and 2500Hz, I could have dancers out in front of my speaker stacks being pummelled by the beat - yet being able to TALK to their friends and have conversations - and be heard.

Nowadays I apply the same techniques to national TV show mixes. In both cases, nobody was the wiser, but they did have more fun...

One more thing: the ear never cuts (like the eye blinks).

From birth to death it's one long, uninterrupted soundtrack, no cuts, only dissolves. Anything abrupt spells danger (unless it's becomes repetitive and proves dangerless, at which point it's tuned out).

Just as a mechanical eye (i.e. lens) can never completely take the place of a human eye, a mechanical ear has a similar problem. (no brain to sort the signal from the noise).
One does that in audio post, filtering out the unwanted, adding the wanted (often with ADR), etc.

This side of post-production is the most neglected of all, but listen up and recognize:

if a picture takes 8 weeks to cut, then the feature film rule of thumb is that it's going to take at least 16 to 24 weeks to cut the sound track (dialog, effects, practical music, scored music, international tracks, surround tracks) and around 2-3 weeks to mix it (10 hours per minute of final mix), if the audio quality is going to be comparable to the picture quality.

I kid thee not.

Otherwise, we have reality TV.

Cheers
Chris
 
ARRIguy... that was a great question... and we're getting some brilliant answers. I wanted to make the analogy to a camera and our eyes too.

I guess on the positive side of things there are MANY times when a mic can hear WAY better then your ears. Maybe it can't hear as naturally AND better then your ears... but hey, one out of two 'aint bad.

Something else you may want to check out is a large diaphragm studio mic. I've got two NT1000 mics and if I set 'em up just like ears... i.e. 6-8" apart with both of them facing out and forward about 45 degrees... In this configuration these mics DO sound a whole lot like your normal hearing... on STEROIDS. The sound is natural without too much emphasis on anything... even AC noise... but everything is pumped up. Everybody who hears it says the same thing... "I wish my ears worked like that all the time!"

But if you play around with it you'll also see why this sort of setup isn't a good thing 100% of the time. You can hear whispers mixed in with any other noise in the room. I made a mount/blimp thingy to use these studio mics in non-studio conditions... for times when I want 100% of the available information from someplace. Like for backgrounds and stuff... and it's fun to later put on some 7506 headphones and pick out EVERY detail of the location... just like you're standing there.

Now with the type of mics that are giving you this enhanced AC and other unwanted sounds... well at least the OTHER thing they do is give you the option of NOT hearing sounds that you don't want emphasized. When used properly you get the option of choosing what you want to put in your movie.

Think of it this way. How interesting would a movie be if the sound in it was TRULY realistic? When in real life have you been able to hear every nuance of a conversation between two people 50' away? Even when the shot is framed for distance... so that we KNOW the talent is distant... Hollywood still makes sure to keep it interesting for us.

The same goes for all the sound... the next time you watch a movie notice how UNREAL it sounds... and why that's a good thing.
 
"From birth to death it's one long, uninterrupted soundtrack, no cuts, only dissolves. Anything abrupt spells danger (unless it's becomes repetitive and proves dangerless, at which point it's tuned out)."

That is very intriguing... I've never heard it put that way.
 
Great post. So let me ask the question since I am new to everything. The best way to cut the background noise seems to be "get the mic as close to the speaker as possible". Is this correct?
 
Getting the mic close to the subject is rule one because, like light, sound falls off according to the inverse square law (double the distance and the sound is 1/4 as loud). Understanding how sound waves reflect or are absorbed by different surfaces and the characteristics of different microphones is almost as important.
 
You must log in or register to reply here.
Back
Top