VideoHelp Forum
+ Reply to Thread
Results 1 to 4 of 4
Thread
  1. Hello, I recently filmed an interview for a documentary I am filming, and people were being very noisy in the background, and ended up ruining my footage. Is there any program out there that can help with this? I suppose what I am really asking, is if there is a program that can isolate different sounds with in a single piece of audio and let you adjust the levels of all the different sounds, I get the feeling something like that doesn't really exist?

    thanks for any help.
    Quote Quote  
  2. Member
    Join Date: Sep 2007
    Location: Canada
    Search Comp PM
    Only on CSI and the movies.

    So not really - at least not perfectly

    You can try to isolate certain frequencies and adjust them, but there is more often than not , too much overlap between the sounds for this to be truly effective

    If you upload a sample, people will examine it and suggest if something can be done to improve it
    Quote Quote  
  3. ok, I will try to upload a sample. I just did a google search on this topic, and apparently there is a feature on Audacity that lets you remove noise from your track, I wonder if that would work for the sound of people talking?
    Quote Quote  
  4. Member
    Join Date: Sep 2007
    Location: Canada
    Search Comp PM
    Originally Posted by RVStudios1 View Post
    ok, I will try to upload a sample. I just did a google search on this topic, and apparently there is a feature on Audacity that lets you remove noise from your track, I wonder if that would work for the sound of people talking?
    Yes, definitely try it out. Audacity is actually a decent choice (and free)

    Most audio NR algorithms rely on you getting a good sample of the noise only (so you would sample a section where there is nobody talking, just the background "unwanted" audio) . It gets digitally subtracted from the audio track, so ideally only the "wanted" audio remains. However, there are often frequency overlaps with the "unwanted" audio and the "wanted" people talking - so the speech gets distorted if you apply the NR algorithms at higher strengths
    Quote Quote  



Similar Threads