2

Good evening to the Audio Alchemists...

Recorded three subjects, all hard-wired to Sanken COS-11 lavs, into a Fostex FR2LE with a SD 302 in front of it.

Was taken by surprise when one of the subjects presented with a voice like what one might imagine belonged to a 65 year old, chain-smoking, scat-singer---beautiful and wonderfully textured, but requiring a significant boost in gain just to get a healthy signal due to how soft and croaky it was.

Cutting the interview together, the backgrounds are (obviously) dramatically different and jarring when cutting from one subject to the other.

Without access to noise reduction software, is it better to jack up a clip of room tone with gain, and make long, slowly faded heads and tails to Scatman's clips, or use the same boosted room tone to lay underneath the other subjects clips to keep the background noise consistent?

Any production techniques that could have helped me out on location?

Thanks for taking the time to read my query! atheisticmystic

Improve this question
2
  • What software do you have at your disposal?
    – Bluesman69
    Jun 24, 2011 at 12:12
  • @bluesman...greetings o' sage. I use PT8LE with its standard plugins, and I have Soundtrack Pro, and Audacity (although no experience with the latter two).
    – atheisticmystic
    Jun 24, 2011 at 12:39

4 Answers 4

Reset to default
1

@atheisticmystic Both Soundtrack Pro and Audacity have noise reduction plug-ins, so you should be sorted for the type of noise reduction you require (I'm not sure about PT8LE). Anyway, either one will allow you to reduce the background noise on the channel that you need to boost. This will most likely be a more effective solution than pushing up your noise floor.

The module in Soundtrack Pro is very simple to use, but is quite limited in the amount of reduction you can apply before introducing artifacts, so you will probably be better off using Audacity. Even if you have never used Audacity before, it is a fairly straightforward application. Also, general broadband noise reduction is not really a difficult process. You just need to find a few seconds containing only noise from which you can make a noise print. That noise print is then used to create an algorithm that is applied to the entire file.

Don't try to remove too much - just enough to bring it back to the original level. It might be a good idea to mix in a little room tone anyway, just above the original level, as this will help everything to blend.

Improve this answer
2
  • @Bluesman...how critical is the length of the "noise print" section of clip? Using a break between words renders pretty swimmy results even at the lowest setting on noise removal. Is it imperative to find a section of clip that is several seconds in length?
    – atheisticmystic
    Jun 26, 2011 at 16:24
  • @atheisticmystic It does help to find the longest section possible, as this gives the algorithm more to work with, although anything more than 5 seconds will generally not add any benefit and will probably also increase processing time. The beginning and end of the recording are usually the best places to look.
    – Bluesman69
    Jun 26, 2011 at 17:31
2

If it's bearable, I would probably bring up the level of room tone underneath the other subjects to make it consistent. It's not the ideal solution, but it will be far less distracting than the noise floor changing on each shot.

Improve this answer
2

Sometimes when a speaker is really soft a little bit o' compression can help. If you don't have the gear, then eq, no eq then ride the fader, no fader then sometimes you can eq the room out in post. On location nothing is more important than mic placement and gain settings. (cutting 450 hz never hurts either)

Improve this answer
1
  • @Ian... I felt good about the lav placement, but really had to ride the gain on his channel. He was just not agitating many air molecules with his voice. +1 for the compression suggestion, and are we talking like a 1-3db cut at 450hz with a Q of like 1, or something really narrow and deep?
    – atheisticmystic
    Jun 24, 2011 at 12:52
1

It will take less than 30 days to do the work so the iZotope RX demo will be perfect for this job.

Very easy to use, great sound quality, RX is awesome!

Improve this answer
2
  • @AGZFX...Demo version of RX disables saving, so how do you envision it being perfect?
    – atheisticmystic
    Jun 26, 2011 at 16:21
  • oops, I thought it was just time limited, let me change my answer. Buy RX or send the file to someone who has it. Lots of folks on SSD have it.
    – AGZFX
    Jun 26, 2011 at 16:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.