Users browsing this thread: 4 Guest(s)
Removing garbage noise from a sound stream
#1
I read somewhere on the internet that by choosing "Raw Data" from the Import list, you can use Audacity to access certain sound files from games, so I used it to open a soundpack from Cel Damage. The soundpack in question is TalkToMe.dat, and includes all in-game voice clips. Since VGS Converter does not support this file type, I had to use Audacity to look at it.

After making many adjustments, I got it to the correct pitch and speed that the voices are supposed to be at. Unfortunately, there is still lots of garbage noise mixed in, making the voices sound somewhat garbled. No matter what I tried, I could not remove the noise at all. How do I go about removing it?

Here is a link to the result if you would like to have a look: https://www.dropbox.com/s/9viqyd1xb9v075l/talkToMe.wav (and yes, that is Charles Martinet that you hear from 7: 36 to 9:40, for anyone who is wondering)
Thanked by:
#2
There's no easy way to remove this noise, if at all. The only way I can think of is by modifying the waveform itself, which isn't really an option (both because that's ludicrously tedious and difficult, and because it won't produce the original sound). Unfortunately you'll have to try and find a way to properly extract the audio.

If you could upload the .dat, I'd be willing to take a look though. Can't promise anything of course, but it doesn't hurt either.
You may have a fresh start any moment you choose, for this thing that we call "failure" is not the falling down, but the staying down. -Mary Pickford
Thanked by: Mystie
#3
There's no easy way to remove this noise, 1. Because the hissing and humming dominates the vocals in volume. 2. Any editing to it will also effect how powerful the vocals will sound, making them sound like they where out of breath and artificial. 3. This is a mono type file, unless this noise was a stereo based file with mono type vocals mixed in, then we might have a chance in retaining the vocals, but as it is now, it's a no go. You'll have to find another way to rip it with either no noise or with the vocals dominating the noise.
I've made my own Vaati Battle style, I used the original sound effects in the Game, but I used songs I retained from the English Dub of Kirby Right Back at Ya, enjoy.
The Legend of Zelda Four Swords The Great Wind Mage Vaati
http://www.youtube.com/watch?v=qbqtfxYMpZE
The Legend of Zelda Four Swords Adventures The Great Wind Sorcerer Vaati
http://www.youtube.com/watch?v=X66EUcSdOks
Thanked by: Mystie
#4
Thanks guys!

I'll go PM puggsoy with the .dat file right now.
Thanked by:
#5
Hmm. Sorry, unfortunately I don't see anything helpful and can't find any info on it. At this point your best bet might be trying to record ingame, but unless you can disable all other sounds that's probably not really that helpful.

Sorry! Hopefully you can manage to get them one way or another.
You may have a fresh start any moment you choose, for this thing that we call "failure" is not the falling down, but the staying down. -Mary Pickford
Thanked by:
#6
It's fine! At least you tried! Wink

I'll record them from the game itself then, and that'll be extremely easy since it's one of those games that lets you adjust the volume of the music, SFX, and voices. However, since Dolphin cannot emulate this game to save its life (lighting and shading are completely absent, emulation is slow, and during gameplay you will constantly fall through the ground and die), I'll have to record from the Xbox version. (I own a physical copy of the Xbox version, and an ISO of the GCN version)

Don't expect to see any voice clips from the cutscenes though, as all audio in the cutscenes are all part of the same track. Turning off everything except the voices (or everything except the music) results in dead silence, while turning off everything except the SFX results in music, SFX, and voice. Strange, I know.
Thanked by:
#7
If that's the case, it's likely that the actual audio of the cutscenes are stored as one part of audio data (i.e. the voices/music/sfx for cutscenes don't actually exist separately). So in that case it should be fine to just record them like that, and upload them to TSoR as "Cutscene Audio" or something.

Of course, if you want the voices for your own purposes, that's not really useful.
You may have a fresh start any moment you choose, for this thing that we call "failure" is not the falling down, but the staying down. -Mary Pickford
Thanked by:
#8
Oh really? I thought that having music in submissions to TSoR was against the rules.
Thanked by:
#9
Oh yeah, that kind of flew over my head Tongue You're right, music can't be submitted, even if it's in the background for cutscenes. My main point though, was that it's likely that voices in cutscenes can't be separated from the other sounds even if we could extract audio directly.
You may have a fresh start any moment you choose, for this thing that we call "failure" is not the falling down, but the staying down. -Mary Pickford
Thanked by:


Forum Jump: