uploading sounds to SL

Well, I never noticed that we do not have a sounds forums for issues with creating and uploading sounds....

Anywho...

my question is... how loud does a sound (wav) file have to be, to be loud enough when uploaded to SL?
I have decided to try my hand at gestures, making my own sound files with audacity, and lo and behold after about 200L later with uploads, I just can not seem to get it loud enough (finally remembered to use the preview grid so as to not waste lindens)

I have it super loud in audacity but when uploaded, comes through soft and low...

any advice, suggestions, comments, help, anything...

Thanks

Can't help with what volume you should be using in Audacity, but just wanted to ask (because sometimes the obvious is *so* obvious that it gets overlooked) - have you checked your individual sound settings in the viewer to make sure they're not too low?

yes that was the first thing I checked, I maxed my sounds

I also asked others if they were low to them, (with various sound settings) and they were, as well as compared them to other sound files... I just wish I knew the best way to make them loud enough, without distorting them too much

blech

(why oh why isn't there a sound forum grrr)

Sounds that are normalized for standard CD levels are about as loud as you can go. Any louder, and they'll end up clipping at input on most audio devices.


SL's got two strikes against it for audio levels. First, no matter what you do, the master output level for SL will always be far below that of just about any other program. Why this, I have no idea. It's just the nature of the beast. SL is quiet. I haven't actually measured, but I'd estimate SL's maximum output at a good 6-10 db lower than that of, say, Windows Media Player. There's really no way to overcome a gap that big by trying to raise the source levels. You'll just end up clipping it.

Second, sounds do fall off over distance. The further your camera is from the audio source, the lower the level you'll hear. If you want to, you can compensate for the distance falloff by adjusting your preferences, but if you overdo it, you'll ruin the sense of immersion in the world, as sounds will no longer seem realistically spatialized.

I can't help with the uploading part, as my own attempts have had mixed results. However, gestures are special in that you can make those as loud as you want. All you have to do is put multiple instances of the same sound in the gesture with no wait time between them. If you use more than two you can get an echo effect on the first use, but most of the time the effect is just the sound is louder. It usually only takes two plays of the sound to make it loud enough anyway.

I'm well familiar with this problem, given that I've had to upload sounds for my avatars etc on many occasions. Audacity is a great sound editing program, but I've found that upping the volume of a sound in Audacity does NOT produce a sound that is sufficiently loud for my needs in SL. It's weird, but I have no idea why that is. What I do is to get the sound the way I want in Audacity, export it in the appropriate format, then open up that wav file in the Windows Sound Recorder app. In the Effects menu, the first entry is "increase volume by 25%". I do that repeatedly until it's as loud as it will go without noticeable distortion; I then save that file and upload it.

I haven't the foggiest idea why that works, but it gets me much more usable volume on sounds uploaded to SL than anything else I've tried.

@Maya: That multiple instance idea is brilliant. I wish I'd thought of it. Thanks for sharing it.

@Nimbus: It's been a while since I've dabbled with Audacity (Adobe Audition is my audio editor of choice), so I can't say for certain, but it's possible its clip threshold is lower than standard. It's also possible that the threshold is configurable, and maybe just the default is low for some reason. Either scenario would explain why you're able increase the volume of a sound in Sound Recorder beyond where it would normally start to distort in Audacity.

Torley has a great video about this:

http://www.youtube.com/watch?v=QBVmFafFatE

thank you all very much

I will try the sound recorder thing, when I do another set of sounds. At this point I think I got them usable LOL

I might also try the doubling up of the sound file in a gesture, see what that does (that might be how others are doing it too, dunno)

Chosen I know about the falloff from a distance, that was not my issue.
With my cam right at my avatar the sound was still very low, people would come stand right next to me, still very low.

So my aim was to make them loud enough for everyone to hear at about mid volume level that were relatively close to the source.

I think the suggestion of using the windows recording is something I will try.

I had a look at torley's video, it doesn't explain about volume though

all the steps he outlined I already know how to do (I am just stuck on getting it loud enough, and he skips past that very important fact, that normal volume in Audacity will be very low in SL)
I also wish we could preview sounds like we can with textures, thus confirming if they are loud enough....

Does anyone know whats the length in the wave format to making a gesture with a sound recorder? Ive tried to make one but when i try to upload it into SL it tells me that the that the file does not appear to be a supported sample rate must be 44.1k. So im not sure how i'm suppose to create this gesture. Ive done it once along time ago but i cant rememeber what i did. Can someone please help me?

I use Audacity to up the volume, and I goose it to a little ways above the volume where it says it will clip. I've compared the sound quality of the original and the louder sound, and I've never noticed any difference.

However, I've just been doing voice, I've never tried it with music, where I imagine it would be more noticeable.
.

I've not found this to be true.

I haven't done sound work in SL in a couple of years, but I seem to recall Audacity has an option to scale the sound to maximum automatically. You need to use that.

be careful with any of the compression-style effects. overuse can negatively impact nuance.

I'm by no means an audio "pro", but I have had the same issues as well.

Chosen is correct; the volume levels in SL are significantly lower than other programs. I don't know if it is the spatialization or the sound engine itself, or a combination of both. However, if you want to be able to hear the sounds at a reasonable level, you need to normalize then amplify them just under the point where they start to clip before you upload. Even still, they won't be nearly as loud as they are locally on your computer because of the problem described above.

I've used both Audacity and SoundForge, various versions, and I've noted no difference in mitigating this problem. As such, I don't think it has anything to do with the program used to create/modify/mix the sounds.

The gesture trick with multiple sounds can be used to get around this problem, but it only works for gestures. I've tried to do the same thing with scripting using llTriggerSound, but with mixed results.

There are some things which seem to not be possible in SL at the moment, related to sounds. An example: church bells. The spatialization falloff pretty much means you cannot generate any sounds which even remotely resemble the carry of something like bells, no matter how much you amplify/tweak/trigger multiple copies/etc.

Rhaorth Antonelli,

Sound Files uploaded to SL have to be 9.9 Sec or less in length and PCM Uncompressed 44100 hz, 16 bits, mono and a .wav file...

I use a program called Pyro which will clean up a sound file and allows the loudness to be increased some without distortion...It will generally bring most sounds up to a reasonable volume when uploaded...

Adding the sound multiple times in a gesture works well to increase the sound level in a gesture...It can get strange in a laggy sim LOL...

You can chain sounds together using the wait step in gestures to get longer sounds...However, again Lag will sometimes effect that also...

If I can help...Just IM me in Game...Usually on 5 am- 9am SLT

Kurt Ludd "The Gesture Maker"
Try a FREE 12 Gesture HUD at http://slurl.com/secondlife/Malra/15/57/66

thanks everyone

kurt, the info you outlined I already knew. (Thank you for the offer to help though, might take you up on that sometime heh)
It is not how to upload, or how big the file can be (lengthwise) or chaining them or doubling up, it is basically how to get the sound loud enough that it is normal in SL, without distorting the sound (I think they call it clipping)

I tried the suggestion of using the windows sound recorder and that seems to be the best option.

I clean up the sound using audacity and the noise removal option, works perfectly for me.

I tried the doubling up the wav in the gesture and that left me with unsatisfactory results.

So far, the best solution (for me) was the sound recorder suggestion for increasing the volume. Thank you Nimbus Rau for that suggestion.