Normalizing audio help

[1031982]

What hardware and software I have to work with are as fallows :
SoundBlaster Audigy 2 ZS Platinum and the Logitech Z-5300 speakers. Properly calibrated
GoldWave, and some micellanious free programs.

Here is the thing, I have a video with some interesting audio. It has a few spots where it's loud/normal, and the rest is extremely quiet. I want the quiet parts to be louder, obviously, but the loud parts are a legitimate part of the sound. The other thing I want is for it to be seamless.
Manually lowering the parts will take a good week, at least because there are spikes throughout. Using GoldWave's auto features causes a lot of things to sound muffled and/or obviously edited.


The audio is encoded with WMA , mono. I plan on re-encoding it with a larger bit rate to help keep the loss to a minimal. I know there will be loss no matter what.
Any ideas?

[redwudz]

I use the freeware Audacity most times for normalizing. Not sure if it handles WMA audio, but check the new versions. It handles quite a few formats. Audacity's normalizing doesn't seem to cause problems in my uses. Of course if your file is really messed up, it's a bit hard to restore it to 'normal'

With audio, re-encoding doesn't have as many negative consequences as video. I rarely notice any difference unless I really mangle the file with excessive filtering.

[t0nee1]

Yes, the latest Audacity version 1.3.7 beta imports/exports WMA audio with ffmpeg.

[1031982]

Probably should have mentioned that I am going to be using a wave export.
It just makes things easy with applications in Windows.

Anyway, that doesn't work. I get the same issues.
These pictures are from GoldWave. From top to bottom, they are the sound file as a whole, a loud part, and most of the rest of the file. This is the original audio, before doing anything other then working with it as a wave.

joined.jpg

[ZQX]

Normalizing just raises the highest peaks of your entire audio to maximum decibels without clipping - it doesn't "equalize" the volumes across the entire wav. So "softs" are still soft and "louds" are still loud. It's just all shifted up higher, as high it can go without the highest/loudest bits clipping.

What you want is compression.

[1031982]

Well, trying that doesn't do much. At best I get a .5 db change.

[johns0]

In goldwave did you try effect/compressor/expander/boost quiet parts?

[1031982]

Yes, but again it becomes muffled. No clue as to why, as I am not changing anything else, and if boosted manually there is no issue.

[ZQX]

Tweak your compression settings. Use a higher ratio, like 4:1 or up.

[netmask56]

If you can use VST plugins this site lists quite a few free programs

http://vstplanet.com/Effects/Bundle2.htm

[1031982]

Thanks Everyone. I used Goldwave to edit it in the end, but it's a lot better.
I had to use a ratio boosting the quiet sound by 300%, run the reduce peaks, and finally normalise it again.
Nothing got muffles, loud parts are still loud, and the low parts are at a good volume range with clarity.

I know enough about audio editing to do some things in Goldwave, but this was annoying as it was from a professional source. Just horrible mixing IMO to let this go by. I do some very basic VHS to DVD conversion of home movies, and I spent a good 7 hours on one audio clip for it to be equaled out, no background noise, and keeping the clarity. I would think that would be minimal for any company.