Ricebug on 11/5/2013 at 12:59
I was just over on the FM forum, and I see that some folks were bitching about a problem that comes up every now and then, i.e., sound files that are hard to understand. Specifically, conversations, where folks don't have access to professional equipment or voice actors.
I'd like to offer whatever help I can in this area. I have a recording studio in my home. Barring sending you a Rode mic, I can take your file and run it through a hardware- or software-based processor and clean it up. PM me if you're interested. Having been involved in studio recording most of my life, here are couple of tips for those home-grown recordings:
* What sounds good on your system may not necessarily sound good on all systems. Throw it on a CD and play it on your car stereo, a boombox, a neighbor's stereo. This will give you an indicator if there are problems with it--tinny, boomy, nasal, etc.
*
* Record the voice so it sounds like it was done in a sound booth--no room artifacts, no reverb, no wife banging pots and pans downstairs. A heavy blanket thrown over the voice actor and microphone can work wonders when you don't have a $20,000 booth.
*
* Use effects (reverb, chorusing, etc) like salt on food. A little is good. Don't overdo it. ALWAYS save your original untouched recording as a "dry" file. Keep it in a safe place in case your effects didn't work out so well.
*
* Always do your recording "flat." IOW, don't add bass or treble to make it sound good while you're doing the recording.
On the other side of the issue are those players who like to run Thief through equalized and reverbed sound cards "because it sounds cool." Sometimes they have trouble making out those homemade conversations.
gigagooga on 11/5/2013 at 13:32
Quote Posted by Ricebug
I can take your file and run it through a hardware- or software-based processor and clean it up.
Traktor?
I use that myself :P
Sliptip on 11/5/2013 at 14:05
Excellent suggestions Ricebug! :thumb:
The only thing I'd add to that list is that I find most voice recordings found in FM's would benefit from some compression! I find it really helps the dialoge sit above all the background sounds. But as Ricebug pointed out, only if your wife isn't banging pots in the background! :)
Ricebug on 12/5/2013 at 12:56
gigagooga: Traktor is for DJ work, but I suppose it has some fancy tools. Mind-blowing technology if you've never seen it used live.
Sliptip: I religiously compress (almost) everything. For those who don't know what it does: sound files, especially vocal parts, have loud parts and soft parts. A compressor will "squash" the sound to bring the louder parts down to a level where it pretty much matches the rest of the waveform. Good compressors also "expand" (called compressor/expanders) to bring up soft parts and bring down loud parts. The result is a nice, smooth, vocal part that plays at pretty much the same volume all the way through.
Child Of Karras on 12/5/2013 at 22:47
Compression and equalization are the only things that should be done to voice files. Also: compare them to the standard voice files. If they sound like the same engineer has worked on them they will fit well in the mix ingame.
gigagooga on 13/5/2013 at 05:53
Quote Posted by Ricebug
gigagooga: Traktor is for DJ work, but I suppose it has some fancy tools. Mind-blowing technology if you've never seen it used live.
It's very handy for mastering, and with controller to go with the software you can re-record lines very fast.
ocn on 13/5/2013 at 06:43
Sound is so incredibly important for immersion. This is an offer we cannot refuse. Thanks, Ricebug.
EDITED ADDITION: Another tip would perhaps be to try and situate important conversations in areas where voices recorded under different conditions can be made to sound more like they're in the same locale. It always throws me off when they sound too different. Through the use of for example cam vator sequences, the conversation can almost be viewed as a seperate artistic entity where it is more forgiving to suddenly add subtle effects (like reverb if it's in a hall for instance) too make things more coherent.
ffox on 22/7/2013 at 10:53
Apologies if this a bit OT. If a large number of sound files (eg 45) are pointed to by one voice index, will they all be played sooner or later? If not, how many should be junked?
Yandros on 22/7/2013 at 11:35
Well, it depends on the schemas I suppose. Theoretically if all the files are included under at least one schema, they will all be played eventually. Although some schemas are oriented around events which don't occur very often, making them much less frequent than, say, your idle comments and mutterings.
john9818a on 25/7/2013 at 17:32
When I first started using my Creative Labs Fatality card the 3D effects were on by default. I couldn't stand the effects because it sounded like I was playing inside a church bell!
I've noticed a problem with using OpenAl. With the regular hardwar acceleration and Environmental Surround (EAX) sounds/voices heard behind closed doors were correctly propogating from behind the door, but with OpenAl the muffled sounds are propogating from the center of the room I'm currently in. I know it isn't a room brushing issue because it happens to me everywhere.
Sorry to be so off topic but my Motorola walkie talkie has a compression/expander feature for use in noisy environments.