Uploading Sound Files

[Prokofy Neva]

I heard that the Lindens were moving up from a limit of 9.5 seconds on sound files to 30 seconds. But I can't seem to find an official notice of this. Did they?

Meanwhile when I was trying to upload some poetry and a tune (wav files) in recent weeks, I got an error file that says the files cannot be longer than 30 seconds in length -- as if that is indeed their new maximum size.

Since the file in question was exactly 30 seconds, I couldn't figure out why this didn't work, when it was in the proper format.

So I tried 29 seconds -- that didn't work, either. So it's still on 9.5 is it?

I have gone through many hours struggling with Audacity that work, and I will tell you the steps I have learned to take in part from Whirly and in part from trial and error, in case you know anything different:

1. Change the file from m4 to wav files using online converters (if from a YouTube video, for example).

2. Make sure you set your project in Audacity to 44100 Hz or SL can't accept it.

3. Go to the top left of the screen on the file and push the arrow down to "Split Stereo to Mono"

4. Delete the second copy of the file.

5. Select the whole file and go to "Effect" and select "Amplify" -- I do this because many files I have uploaded, especially of people reading poetry or literature, are just too low volume, even pushed to "1" on the player script. I select in the range 11-13 because beyond that, it gets so loud it is distorted. You can then go back on the player script and adjust down from "1" to ".9" or something but I was informed that loudness or too-softness has to be fixed in Audacity before the upload. This is not ideal. Maybe someone has some more refined suggestions.

6. Now you can go in several directions:

   a. the arduous (and I find painful and error-ridden) of hand-cutting the files into 9 or 9.5 segments using "Edit" and "Clip Boundaries" and "Split". At least in doing this, you can get rid of noise if you took something from a noisy RL room or something, and do other things. It's pretty hard for me, however.

   b. the also error-prone method of using "File" and "Import" and "Labels". Here you need to first make a notepad file with the time stamps listed in it then import that file and that will make the job chop up automatically according to the time stamps you set.  You have to check and double check that the list you make doesn't have any math errors in it that mess up the whole project -- so it's 9.5, 18, 27.5 etc. in a table form (and for some reason online I found examples of this showing the tables doubled up, i.e. in two rows on the notepad file). If you have made an error, you get a message that Audacity can't read some of the labels; it will still work but when you try to listen to it, it will stop and skip at some file, then return to the first file, causing you to have to re-do everything and pay the upload fees again. It amazes me how you can make mistakes on this again and again even using a calculator because you get weary typing.

  c. a much, much easier way I just stumbled on (I am a slow learner) where you go to "Tools" and "Select Regular Intervals" which makes the "labels" or cut-up files for you automatically, obviously with no math errors. The one pitfall here for dummies, as I can tell you from experience, is that you have to roughly figure out how many labels you will need, then listen to then, and possibly have to go back and re-do it, and then make sure that whatever last file you have isn't 'the rest' of the file, i.e. more than 30 seconds, but only 9.5 seconds. SL won't upload files chopped up with no sound on them, or, as noted, files that are too long. On some scripts, if it is under 9.5, you would have to note that it is, say, "4" or it messes up again in terms of skipping and starting over.

This is a pretty beastly business -- and costly, with those $10 uploads. You have to really want to do the thing, like add poetry being read aloud to a scene or something. You can also just put a URL linker into a prim and it will take someone who clicks online to a YouTube or a literary page. Of course, that isn't as "immersive", but ask yourself: how many people are going to click on a prim in SL to listen to Wallace Stevens reading "Idea of Order at Key West" or the Ultimate Spinach's "Mindflowers," a song lost in time for sure.

Answer, as I can tell you from making these builds: not many. Do not expect to make back your upload cost in tips or sales. One issue is that if there is a media stream already running a sim, it drowns out other things making sounds like this. You have to make sure your parcel doesn't have media, and/or if does, instruct the visitor to turn off the option "steaming media' in preferences. How many will do that? Again, I have answers for you from running these sites for years: not many LOL.

I mean, if you have a burning desire to listen to Otava Yo (which I do) while sitting on your SL sim, why can't you just...go to YouTube and turn it on and listen to it on your headphones. You can then even leave your sim and fly around the ocean and continue to listen to this tune stolen from Caucasians by Russians who can't dance (although the video is pretty cool).  Why sit for HOURS chopping it up, even if to 30 seconds if the Lindens have that now, listening to it inevitably skip or short out due to lag or whatever, when you can...just go to YouTube. Soon you will wonder: why sit in SL when I can sit in my RL living room with my RL friends lol?

Of course, there's the issue of copyright, too. It's one thing to upload these files to make a reference inworld, which is no different than a link on a web page, really. It's another to make up a box of it and sell it, unless it is in the public domain. The reality is that the cost of uploads are high, the work painstaking and poor quality, so I think few people bother with this.

And really, the era of listening to a discrete song as some kind of discrete entity that you like and learn and play over and over is past, no? At least, I find that with my kids. Most young people have "playlists" these days and streams from some device or another that...streams. They think in terms of "my wake-up energy playlist" of 10 songs, not "this one song I like".  You can play a song again by clicking on it on YouTube or Spotify -- but since you have access to zillions of songs, why not listen to another one? I don't think people listen to a song as a separate thing the way they used to.

So, you say, there is media on a prim! Put the YouTube into media on a prim or a YouTube player. Answer: these players often don't work because of issues between SL and YouTube of the type that are between Twitter and Flickr which caused the Lindens to stop offering this option of posting to social media from SL. Media on a prim is ok, but clunky. It has a way of jerking your view to itself. Also, you can't keep it running. Even if you have autoplay turned off on your own YouTube account on your computer, in SL, it will autoplay. You can't get it to play and stop or just play on a loop. No, the URLs people have devised to make things loop just don't work (any more).

So in that sense, if you task is to create an ambiance on a build for an expo (as I'm doing for RFL Home & Garden now or as I have done at the SL birthdays in the past), and you put up video on a good YouTube player (there are some conscientious makers who keep up with the frustrations) or "media on a prim," chances are it will stop playing or some other completely different song will be on when you come back hours later.

So in that sense a prim box with the audio files in them chopped up, that play with a loop script or on/off script is just that much better. How many visitors will care? Answer: not many.

 

Edited February 27, 2020 by Prokofy Neva

[Whirly Fizzle]

A recent viewer change added support for up to 30 second sound upload, BUT as far as I'm aware the serverside change needed to support this hasn't been implemented yet.
As far as I know, 30 sec sound upload is coming soon (tm) but I haven't seen any official announcement yet when this will happen.

It's also possible that only Premium or Premium Plus accounts will be able to upload sound files over 10 secs in length.

[Liam Metaluna]

the labels i used to import with 32bit Audacity don't work with 64bit Audacity, just saying.

glad I kept my old computer.

[Coffee Pancake]

On 2/27/2020 at 6:34 PM, Prokofy Neva said:

.....and costly, with those $10 uploads.....

Unlimited indefinite hosting and delivery of audio files for L$63 ($0.23 US) per minute.

[Coffee Pancake]

Removing duplicate post. Not my fault.

Edited February 25, 2021 by Coffee Pancake

[Prokofy Neva]

6 hours ago, Liam Metaluna said:

the labels i used to import with 32bit Audacity don't work with 64bit Audacity, just saying.

glad I kept my old computer.

I recently completely re-did my labels for the 64bit Audacity, and while they do generate error messages, they do work, and I do manage to upload them. Not always, for reasons that often mystify me, but usually. I put them at 9.5.

[Mr Amore]

Forgive the revival of an old thread, but has there been an update on uploading 30 second sound files. Is it still on the horizon?