Best way to check processed audio
 

Hi all,

I am faced with a dilemma due to various listening devices. When I make a video with an interview and music in it, the sound always sounds different on different devices. What I am specifically referring to is the fact that on some speakers the music seems too low and the interview too loud or vice versa depending on the device, sometimes I almost can't hear the music but the interview is fine, but on another listening device I can hear both just fine. How do I figure out the optimal levels? My monitoring headphones (Sennheiser HD 280 pro) are very sensitive so I hear things really well but then I listen on another pair of headphones and things seem too low. How do I go about that? I almost want to ask what levels should the music be in relation to the person speaking in the interview but I am not sure if that's the right question.
Can anyone help?

Re: Best way to check processed audio
 

Generally it's not possible to find good final mix levels for speaker playback when monitoring with headphones.

Nothing against the Senn HD280 cans. They're great, and I use them all the time. But not for mixing in post.

On speakers. Neutral speakers that don't overstate bass, like most home stereos. Neutral like studio monitors.

There are some boxes from M-Audio and Yamaha that only run a very few hundred USD. But the price goes up fast, with some *very* good pairs in the $1200-2400 range.

Any of those will get you very close in most editing environments, but is really just the beginning of a discussion about acoustics of your editing room. The goal is neutrality. Then, with good skills and practices, you'll be mixing (and compressing, and equalizing, and other filters) to a reference that provides good results across most distributions.

On the other hand, if you're mixing for headphone listeners, the HD280 are very good.

The first step in using prerecorded music from a production music library is to turn it down by 12 to 16db, depending. Then you may be using volume envelopes to bring it up or down further in relation to dialog. Not to mention EQ on the music and dialog, and dynamic range compression of dialog becomes almost mandatory for good hot mixes. The general rule is that dialog *must* be foreground, and only one source can be foreground at any particular moment.

P.S. I feel sure that there have been many discussions on this forum on monitoring... you could search on monitors, speakers, m-audio, genelec, blue sky. Search is your friend!

Re: Best way to check processed audio
 

I agree with the above. Use at least some moderately good full-range monitor speakers. I like the older JBLs, like the 4408 as a minimum, or the 4410 or 4412. "Old school" but tried and true.

You might also want to check the mix on some small "home quality" speakers, even some mid-line computer monitors (i.e. NOT the $10.00 range), to be sure the dialog still comes through cleanly.

And last, but not least, be sure to switch your system to mono from time to time, as this will affect the sound of the mix, and will also reveal any phasing issues with the mix.

Happy Trails!

Re: Best way to check processed audio
 

I agree with the above as well. You have no control over the PB system once it leaves your hands. I always check a 'mix' on multiple speaker systems..this includes full sized studio monitors, active near fields w/sub, small near fields (Auratones in my case), and an 'average' everyday mono TV set. Minor compromises are not uncommon.
I would not make any mix decisions based on headphones, except maybe to make it easier to detect extraneous clicks, bumps, ect.

Re: Best way to check processed audio
 

Remember that intelligibility in a mix does not depend solely on level. It is also a function of frequency content. If the music is full of bright midrange content (e.g. a wailing saxophone) it will damned well obscure the voice, even if the music level is lower.

And if the playback speakers are lacking in mids, intellibility will suffer, so you need to check for that when you play the track on different monitors.

OTOH if you boost the mids indiscriminately, you might end up with dialog that's downright painful on some speakers.

(This is why THX was a good idea in movie theatres, back in the day. It insured that the playback system met certain specs in terms of level and frequency response, and even insured that room noise and bleed from adjacent theatres were within certain limits. Once all the theatres were reasonably uniform, the mixer could be confident that his mix would sound good in any theatre. You, on the other hand, are faced with exactly the opposite situation ... no standardization whatsoever.)

Re: Best way to check processed audio
 

Some of what you describe sounds very much like an issue with mono-vs.-stereo-vs. -surround sound playback settings. I will add that if you know the likely ultimate playback environment, be sure to try it there.

Also, age of the listeners ears can be a factor as well. When young my hearing was very different and much better at discriminating between foreground and background sounds. That may be a factor to consider in you mix (not my ears, but your clients' ears <G>).