Audacity use to "fix" overmodulation

[reboot]

Have a bunch of files, and when played, the audio is way overmodulated, to the point of noise almost. It's near impossible to hear voices, everything muffled, as if in a big can, and very loud.
Can I use Audacity to import the audio from an .avi? (How?)
Can I use Audacity to reduce the volume, equalize, or whatever to "clean up" the audio?
Using Audacity to export the finished product to .wav, then using TMPGEnc (or Canopus, or Mainconcept) to encode, any forseeable problems with sync?
I didn't find any guides on this, but I'm downloading Audacity's PDF manual now. (but we all know how much one loves reading the manual)

[disturbed1]

Can I use Audacity to reduce the volume, equalize, or whatever to "clean up" the audio?

Any audio editor will let you reduce the volume, and equalize. Some just have better methods of doing it, while others use plugins.

Demux your AVI into video and audio streams, import the audio into any wave editor and play around until it sounds the way you want it to.

My tool of choice is Sonic Foundry's Sound Forge with Waves Production Bundle plugin set.

[au7usa]

Some sounds can be so loud that your decoder can't handle them. This is called clipping.

I can't remember, does Audacity have a Normalize function?

TMPGEnc has Normalize, hit the "Settings" button in the lower RH corner.
The TAB marked "Audio" is what you want. On this TAB is a checkbox with the words "Use Audio Edit". Check the box to enable the function, and then 'click-on' the button "Setting" thats right next to it.

See the "Normalize" button? Hit the button, and enter a value (Scale is 0 to 100), and TMPGEnc will attemp to determine just how loud your audio is. (Notice you can also fade audio in and out from this tab.)

Click OK to save settings for this TAB. Then click OK again to save settings.

Now just encode normally. (try a test file first)

*Zero to One Hundred is the range your decoder can deal with without clipping anything off. (almost)

[MOVIEGEEK]

You could just use TMPGEnc Audio Edit and set the volume to whatever you like then encode,my guess is the audio has been clipped and the quality will be poor even if you lower the volume.

[reboot]

disturbed1. Great! Send me a copy of Sonic Foundry, and I'll use it. (or just send the money to buy it)

au7usa:

TMPGEnc has Normalize, hit the "Settings" button in the lower RH corner.
The TAB marked "Audio" is what you want. On this TAB is a checkbox with the words "Use Audio Edit". Check the box to enable the function, and then 'click-on' the button "Setting" thats right next to it.
See the "Normalize" button? Hit the button, and enter a value (Scale is 0 to 100), and TMPGEnc will attemp to determine just how loud your audio is. (Notice you can also fade audio in and out from this tab.)

I would love to do just that, however MY version of TMPGEnc doesn't have the "Use Audio Edit", or anything like it.

Moviegeek: See above

Would still like to know, can I import audio into Audacity directly from avi? (I haven't had any luck so far).
If I can extract audio, I would hope there's a normalize function!

[feenix]

No Audacity will only import wav,mp3 or ogg. You need VirtualDub or TmpgEnc to demux your audio from your avi.

Oh and yes - Audacity does have a normalise effect. It's easy to use - never even read the manual (didn't know there was one!)

[reboot]

Being the lazy sort that I am, I got OSS Audio Extractor (to go along with my OSS Video Converter Pro) and sucked out the .wav file.
I still haven't figured out how to do that in TMPGEnc or VirtualDub, althought it's probably too simple. (any hints?)
Imported the .wav to Audacity, found normalize...now...what sort of settings should I be using to get the cleanest output?
I could probably play with it for another day or two, but a few ideas wouldn't go astray here

[feenix]

Can't really help you with the settings - your'e the one with the file and only you can hear the results! Just play around with the normalise settings until it sounds as good as it can. By the way ,you don't have to export to do a test play so it's pretty quick to try a setting and test the results.

To extract wav with TmpgEnc open your avi and click on the file menu and Output to File > Wav.

With VirtualDub open avi file and Save Wav.

[reboot]

I messed around in Audacity, and ended up with garbage.
The normalize setting is NOT adjustable. It's preset at -3db.
The "noise reduction" is a joke.
Anyhow, to make a long story short, I used Canopus Procoder, with it's normalize and lowpass filters, and ended up with a VERY nice video, all in one step, without having to extract the audio then remux it all.
That's one of the problems with having too many options, I get stuck in the most complicated procedure, when I should know there's an easier way
BTW, a normalize setting of -15db worked perfectly. I just have to turn up the volume on the TV a bit more than usual, but it's crystal clear audio.
Now if I could only do the same with some of my fuzzy video...

[feenix]

OK well done. Sorry about that I'd forgotten that the normalise function was preset. I think I must have meant amplify which is adjustable (to -db if you want).

Still you've done it now.

[reboot]

Thanks feenix. The more i use Canopus, the better I like it, except for it's extremely long encoding times. Mainconcept is at least twice as fast, but it's output cannot be modified much, so if my original is of questionable quality, I get a lousy product. Canopus fixes things nicely.

I still would like to know what version of TMPGEnc has audio filters though. Mine doesn't.

[feenix]

Your welcome - did it on your own anyway.

By the way, I have version 2.521.58.169 of TmpgEnc and ,whilst I wouldn't exactly call it editing, it does have an edit function with which you can normalize. Don't know if this is what you were referring to?

[reboot]

I "found" a version of TMPGEnc that has the audio editing bits.
Encoding a 45 min avi now, using the normalize.
I'll let you know how it all turns out compared to Canopus.

[Sakuya]

I also have an AVI with muffled audio. I'm wondering how you can lower the volume that much to get clear audio? I don't understand. Since it's -15, that means it is lower right? Sorry, I'm used to TMPGEnc's normalizing where it's always 0 to 100%.

Any tips you can give me on this? I'm using Audacity and it doesn't seem like any of the filters can make much of a difference. Please help!