Jump to content
Sign in to follow this  
Turbine777

Sound file question

Recommended Posts

Hey gang, I have posted this question in the Adobe forum a week ago and if you're familiar with their forum then you are one of only a select few because good luck finding an answer, the place is a ghost town to say the least. I have called Adobe as well and after 3 hours of waiting I was told my question would cost $80 to be discussed, not answered since they cannot guarantee that, just discussed. The fact that I paid $2,400.00 for the suite didn't mean much I guess. I also posted it in RC's forum since the project is mainly for creating a complete voice set, but guess they either don't know the answer or don't care to answer it.Anywho, since this is the general thread and we seem to have a lot more intelligent people here than any other forum I frequent I figure I'd post it here for some insite to what seems to be a pretty simple question (for a sound genious anyway).Here tiz as I posted on Adobe's site (Note: I actually am pretty familiar with it but as some of you may know, sometimes you have to play dumb to get help and feed the ego's of the more knowledgeable):

Ok, peoples, the title is as terrible as my take on this.I have the ADOBE suite but seldomly if at all, use Soundbooth. I was fairly familiar with Audition (and kind of miss it) but even that, it's been too long since I used it. My problem is probably simple, but I haven't been able to figure out why this is happening.I have a project where I need to take some sound/voice clips and save them as: Compressor: UncompressedSample Rate: 11025Sample Type: 8 I TRIED TO USE A PROFANITY HERE - AREN'T I STUPID!annels: Mono The original files I have are: Compressor: UncompressedSample Rate: 48000Sample Type: 32 Bit FloatChannels: Stereo Here lies the problem.When I use the "Save As" option and set the way I need the file as above (top one) it gets demolished. The sound clip is cut in half and quality is practically killed entirely. However, if I just save the file with just one change at a time, like just change the Sample Type from 32 to 8, it's not to bad, then once again save that file but this time lower the sample rate from 48000 to 11025, again, not bad at all. So in other words, if I use "save as" 3 separate times I can maintain quality (to a point, but much better than all at once).Can someone please tell me the easiest way to do this? Perhaps just the order I should do each "Save As" to maintain the best quality? Or even better would be an easier way to do this in one step, again without ruining the clip. I would ask for a possible explanation (short) to help me understand a tad, but rather not waste anyones time since this is so not my specialty and I am much more a graphics person. I have been at this for a week now and am clueless. I have 1300+ clips to convert this way, and would name my first child after the person that could write me a batch file of some sort .Thank you for your time, it's very much appreciated!~DanEdit: If its easier, I can reinstall Audition. So any and all advice is more than welcome.

i9 10920x @ 4.8 ~ MSI Creator x299 ~ 256 Gb 3600 G.Skill Trident Z Royal ~ EVGA RTX 3090ti ~ Sim drive = M.2  2-TB ~ OS drive = M.2 is 512-gb ~ 5 other Samsung Pro/Evo mix SSD's ~ EVGA 1600w ~ Win 10 Pro

Dan Prunier

Share this post


Link to post

Hey Dan,I do not own the Adobe product and do not know how it handles the audio internally, but here is some general food for thought. First, stereo-to-mono conversion is always a bit problematic and sometimes it's impossible to get a perfect result.It depends on how "mono-compatible" your source file is. Mono is what's on the left side of the signal. If the waveform in both channels (i.e. left and right) is very different, you'll obviously lose a lot information when you convert to mono. To check this, try to turn one side off at a time and listen to your source file. How much of a difference is there (other than the sound coming from the left or right speaker)? Zoom in on the waveform and compare both sides. The peaks and plateaus should look nearly the same on both sides. Does Soundbooth offer a phase-correlation meter? It's the easiest way to see how mono-compatible your source is. Ideally it should always peak near 1 (1 means left and right are identical). If it doesn't or even shows a negative value, you will most likely have problems with a conversion to mono. One thing you could try is to clone the left channel to the right side. The sound will only inherit the information of the left side but will still sound stereo. If you do that, you should get a perfect phase. Reverb and other audio effects mess with the phase, so do not use them before converting to mono. Correcting the phase of a signal involves a lot of knowledge and work, so it's best to get it right from the start. If you are recording your own voice set, always record in mono. Recording your voice in proper stereo would require a 2-microphone setup and since it needs to be converted to lo-fi mono for RC, it's not worth the hassle.Decreasing the bit depth is absolutely the last step you want to to. It's similar to going from 16.7 million colours to 256 colours in Photoshop. A lot of colour gradients in your picture become flat and it will just look wrong or in this case your sound will become flat and unnatural. It's also a step that can not be reversed and will affect all further editing of the sound. Bit depth in audio more or less defines the available space between the loudest point and absolute silence. The more complex your signal is, the more bit depth you want. Load up a piece of music (ideally a busy classic piece) and try decreasing the bit depth step by step, all the way down to 1-bit. At that point it's either on or off. I bet you will hear a significant difference with each step. Finally, there is a process called "Dithering" that adds noise to a signal to make it sound more natural at lower bit depths. You may know this expression from Photoshop, because it works in a similar fashion in graphics. So when decreasing the bit depth look for the dithering option as the absolutely final step.The sample rate works similar to bit depth, but instead of volume it's based on time. Imagine a perfect sine wave. The lower the sample rate, the more "blocky" it will become. Similar to a computer game without anti-aliasing. The sounds being cut in half, might be related to the reduced sampling rate. Better try to do each step individually.So in short:

  1. Convert to Mono
  2. Decrease Sample Rate
  3. Decrease Bit Depth
  4. Dither
  5. Enjoy

Maybe this will help a bit. I never made my own voice set, but maybe we can work together on this. Why don't you upload one of the files, so we can look at it?

Share this post


Link to post

Thank you very much for the detailed and educational explanation Alex! Soundbooth and Audition (more so Audition) are professional sound editors, and yes, do have all the above that you mentioned. As for channels, I have confirmed that both are identical so there shouldn' be anything cut the way it's doing. Basically I get the last couple seconds removed and you can see it in the wave form graph. Again it only happens when I save changing all the parameters at once and never happens when I do one at a time.Audition is much more robost I think. Adobe bouth Audition and then adapted their own cut down version known as Soundbooth. They did the same with Macromedia Flash although other than some action script issues I prefer what Adobe did with it over the years. I can go into a sound file and in either program and do things like eliminate a specific sound or frequency, like for instance I edited a movie clip I had where the TV was very loud and couldn't hear the conversation and was able to remove the TV's audio. It has every graph you can imagine to include one that looks like a thermal image of the sound file in colors per freq (not at home at the moment or would give the proper name), but anyway yes I can see every wave of the tracks and both left and right channel have the same data. I'll try what you said in the order of saves and think this will help me a lot! Thanks again Alex. As to why it does what it does when I try to save all the settings at once is a mystery. Adobe may have powerful software but they also have powerful problems haha.


i9 10920x @ 4.8 ~ MSI Creator x299 ~ 256 Gb 3600 G.Skill Trident Z Royal ~ EVGA RTX 3090ti ~ Sim drive = M.2  2-TB ~ OS drive = M.2 is 512-gb ~ 5 other Samsung Pro/Evo mix SSD's ~ EVGA 1600w ~ Win 10 Pro

Dan Prunier

Share this post


Link to post
As to why it does what it does when I try to save all the settings at once is a mystery.
So, how about uploading a file later? I am curious what the problem could be.

Share this post


Link to post

Well I suppose I could, but if you're not also using Soundbooth, don't see the point. It does things in its own way and I think it's that which is the main issue. I'll email them to you directly when I get out of work a little later.


i9 10920x @ 4.8 ~ MSI Creator x299 ~ 256 Gb 3600 G.Skill Trident Z Royal ~ EVGA RTX 3090ti ~ Sim drive = M.2  2-TB ~ OS drive = M.2 is 512-gb ~ 5 other Samsung Pro/Evo mix SSD's ~ EVGA 1600w ~ Win 10 Pro

Dan Prunier

Share this post


Link to post

Ha, I just emailed you, too. See if Soundbooth produces the same problem with the sample I send you.

Share this post


Link to post

Coolbeans, will do and thanks again.I would like to be able to save all the settings at once, but think I'll just write a batch to do the three steps one at a time if I have to. I might also reload Audition later since I normally do after I reinstall but this time around haven't gotten to it yet.Should be home in about 4 hrs.


i9 10920x @ 4.8 ~ MSI Creator x299 ~ 256 Gb 3600 G.Skill Trident Z Royal ~ EVGA RTX 3090ti ~ Sim drive = M.2  2-TB ~ OS drive = M.2 is 512-gb ~ 5 other Samsung Pro/Evo mix SSD's ~ EVGA 1600w ~ Win 10 Pro

Dan Prunier

Share this post


Link to post

I'll make coffee and stick around then. Damn timezones. At least you are East Coast. :(

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  
  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...