So What Should I Do?

[fthrll]

	Quote:	mabran
	Figured this is a pretty open topic here so here goes: Any simple suggestions would be much appreciated. I am using Audacity as my editor.

Hi Mabran,

R Francis Smith, who uses Audacity a lot for mixing audio drama recently wrote this step by step guide of how he does it. Hope it's of use.

http://rfrancis.livejournal.com/683314.html#cutid1

[Bill Hollweg]

	Quote:	Jack J Ward
	With thanks to John Kyle, I feel a whole lot better about doing pretty much everything in Pro Tools. Just wanted to throw that out there. My big challenge now is to get good speakers so they can help me better with a more neutral sound. right now I feel like I'm not ever getting the mix right. I listen on the computer speakers and its right.. listen in headphones, I have to modify... then later.. it doesn't work on car speakers. I'm just about losing my mind on that. J

I mix with head phones Jack, then <when happy with that> listen in the car when it sounds right on headphones- and some nuances are lost due to car noise- but I listen for the main stuff <VA lines being clear- sfx not too loud or too soft based on what's called for in the script etc> and then go back and tweak again.

Wish there were an easier way or a big red button- but that's about the best way I have found to get a nice balance between Car listening and headphones

my 2 cents pardner!
;-)

[Bill Hollweg]

	Quote:	fthrll
	Hi Mabran, R Francis Smith, who uses Audacity a lot for mixing audio drama recently wrote this step by step guide of how he does it. Hope it's of use. http://rfrancis.livejournal.com/683314.html#cutid1

BRILLIANT Article!

[Jack J Ward]

Thanks man.. I've done both.. but its frustrating because so many of the "whispered" lines or low ambient sound gets lost on cars when I mix them with headphones only.
I've found a website with some speakers that John uses to mix with- studio speakers. I think they might help.
Like I said, I'm shooting for that "neutral" tone because in the end I want my stuff to be heard in cars beamed down from XM Satellite Radio if possible too!

[Bill Hollweg]

See I do both for that reason- I mean some nuances will always get lost in a car <the sublte stuff deep in the background> because cars make noise.

So you can blast it in your car and scare people in other cars <my preferred way of listening to AD and dealing with RAP is to play CONAN or MAD MAX blasting outta my car-LOL> it's the meat that must get thru- VA lines and the big sfx and music.

And in my opinion here- whispered lines should be stage whispered as opposed to actual whispers so they come across loud enough in the mix and you don't tear your hair out having to lower background AMB and sfx and music and stuff just so a whisper can "almost" be heard. Of course i see AD as a stage play with the listener sitting dead center front row <it's also how I approach panning> but that's just me- others may see things differently and that's groovy too.

;-)

Even with external speakers like John- unless you play the sfx of an "INT CAR driving SFX" on a seperate stereo real loud, while you listen to your play back at your house on your computer speakers/sound system- I think some stuff will still get lost. Not everything and most will be heard- whether on XM or on an mp3 player or on a burned CD.

Here's a thought:
Listen to a bought copy of LOTR or HHG or NPR Star Wars- or whatever show is a fave with tons of stuff mixed in the show--- on head phones - sounds BRILLIANT right?

Now take that same CD... Get in your car and drive downtown...Drive 70 on a freeway...Still sounds great and all- but some nuances are lost from that crystal clear version at the house. Heck unless your speakers are real frackin' far apart in a room- it will sound different between speakers and the headphones too.

Just thinking aloud here.
;-)

[ccarlsen]

Hi. This is my first post on the site. I'm an Electrical Engineer by profession who loves to write and make dramas with friends. My background with audio is mostly from peripheral studies for my work. From what I've picked up about speakers and such, you may not have much luck getting your audio to sound good in all places. Car radio speakers are not built for hi-fi. Their output spectrum and distortion is not going to be anywhere near as good as quality stereo speakers. And then there's the issue of the placement of speakers in a room, and the resonant qualities (I think they call it 'presence') is going to vary wildly even in home stereo systems. Headphones have other problems, I hear. Apparently when you put speakers right up against your ear, the ear structure causes a frequency peak right in the middle frequencies, so most good headphone makers put something in the headphone that reduces those frequencies so it sounds more like the speakers on you wall.

I believe the music industry used to squash all the frequencies in popular music into a narrow 'middle' range just so that people's favorite tunes sounded the same in the car as it did on their home stereos. Of course, this was just killing the performance on good speakers, so that they sounded the same as the tinny ones. I think that frequency compression makes the audio drama sound flat. You lose a lot of the emotional content of the voice work. I have found that Levelator does the same thing, so I don't use it for my audio dramas.

So I think you may just be stuck having things sound differently on different systems. On the bright side, most people listening in their cars shouldn't be listening to the audio too hard anyway. They need to watch for the kids crossing the road.

[ccarlsen]

Also, since I'm an electronics geek, I can't resist blathering on about techie stuff. For noise reduction and processing the sound in my dramas, I use Audacity and the Ardour DAW software to edit all my work. I've found that using the noise reduction has to be done with a light hand. The best way I've found for myself is to take some raw voice tracks and run them through whatever noise processing I want to use, then listen back-to-back to the raw clip and the processed clip. It becomes much easier that way to see how much of the 'brilliance' or other kinds of sound quality I'm losing.

I use:

Audacity's noise reduction effect: to reduce background noise ( 15 db reduction or less, listen for distortion or funny ringing)

Compression: (volume or dynamic compression, NOT frequency compression) to level out the different voice levels so all voices are at similar volume levels. Also makes whispers louder. (important for car stereos, loud rooms, etc)

Expansion or noise gating: Noise gates turn off all the sound when the sound energy gets below a certain level. This is not the best for reducing hiss, because hiss during the voice is stil present after the gating, so you hear hiss when people talk and then dead silence. Sounds like a cheap conference phone. What it IS good for is removing little sounds like page turns, breaths, etc without me having to do them myself. But I have to remove noise before I run the noise gate. It's also sensitive to the parameters used for the gate. You need to pick the threshold (well below voice levels) and the attack/decay times(50 - 200 mSec for me) properly for your overall volumes.

I find the order I do them in is important:

1) reduce noise
2) expander
3) compressor (Audacity has compressor and expander/noise gate in the same effect, so you can do both at once, but don't do compression before expanding. Compression varies your noise floor, which makes it harder for the expander to gate out the noise, cause levels start changing.)

[JulieH]

The trick I've found to audacity noise reduction is to get the right sample of background noise. I'm not a techie by any means, but I'll try and describe this:

You get your clip of "background only" alone on a track.

duplicate it, so you have two identical tracks of it.

in the first track, highlight it, and go to amplify. If the number the auto-populates is NOT "50", then amplify by 20.

Now in the first track, you will see sound. Any peaks in there, highlight BOTH tracks (the amplified and the non-amplified) and delete them.

after you remove all peaks, highlight the non-amplified track and go to amplify. if it still doesn't auto-populate with "50", then you have a very noisy track indeed.

Once you get an unamplified track to "50", then you can use it as a sample for noise removal without fear of causing those awful chimey artifacts.

Oh, and if when you take out all your peaks, you're left with less than 5 seconds of base noise track (since the noise removal function prefers at least a five second sample to work from) - copy and paste your reduced sample until you have five seconds. I've even done this when the only sample that FINALLY reached 50 was less than a second long - and it still worked.
8)

[ccarlsen]

Nice tip! Thanks.

[JulieH]

It sounds like a lot of work, but once you get it "down" it goes really fast - and it helps a LOT with avoiding chimey noises and those kinds of artifacts.