Please ID unwanted noise & advise?

I don’t think this is mains hum, as I’ve just installed and tried “de-hum” (a triumph in itself - the installation, I mean :wink: ).

This buzzy noise is present on this cam all the time, so I’d be very grateful for any suggestions to improve it. I’ve already done Amplification (because the owls were far too quiet) and Noise Reduction.

https://www.youtube.com/watch?v=Aau7GP5HzJU

It’s not useful to submit work you already filtered because we can’t take effects and corrections out. So Post a raw sound segment. Under 10 seconds stereo you can post a WAV here on the forum.

https://forum.audacityteam.org/t/how-to-post-an-audio-sample/29851/1

If you had to boost the volume enough, you may be listening to the camera cooling fan or other internal workings. The first thing I thought about when I heard it was you are listening to the noise the building or structure was making. Conduction could be sluicing noise into the camera from a cooling fan 40 feet away by conduction, not sound in the air.

There’s another killer problem with shoots like this. I would tell you to put headphones on the camera and listen live while you made changes to position, location, contact with other things, etc. You may not be able to do that because it never gets loud enough to hear it in real life.

That leaves you with the desperation method. Start a long recording. Announce the conditions and let it run for ten seconds. Change something. Like put the camera in the car and say you did that out loud. If the noise is still there, then the camera is making the noise. If the noise vanishes, then there’s something about the location or your mounting. If you have it running while you moved the camera, you might be able to hear the environment changing.

It can’t take you more than a couple of days to run it down. This is not ha, ha weird by the way. This is exactly how I found a mains hum noise in my studio. I had a bass cabinet that failed to go off when I switched it off. mmmmmmmm

Koz

Thanks for all that, kozikowski, however…

I’m not a movie-maker, I’m an online raptor-watcher. This is a webcam on a Great Horned Owl’s nest on Skidaway Island, Savannah - I screen-capture it (in this case, with Bandicam) 24/7 in one-hour segments to make it manageable, and go thru checking for “events”. The area, the equipment, and the already-frail tree, were devastated by Hurricane Matthew and the stream is now “faulty”. To make matters worse, I’m having to use a weak netbook for recording this nest, due to diminishing resources :frowning:

I record in mp4 because it’s the easiest to edit in my very basic editor/converter, which doesn’t do wav, I suppose I’d have to convert it in Audacity but that wouldn’t be raw any more, would it?

This is the site I’m capturing from http://landingsbirdcam.com/cam-3-view/.

Ignore this post clemenzina , and go here.

The constant whine note can be reduced by applying these notch filters in Audacity’s Nyquist Prompt

(notch2 s 204 10)   
(notch2 s 285 5)   
(notch2 s 408 10)   
(notch2 s 420 15)
(notch2 s 420 15)
(notch2 s 914 15)
(notch2 s 914 15)
(notch2 s 942 5)

nest notches.png

I think you’ll still have to apply noise-reduction after the notch-filters to reduce the vacuum-cleaner-like noise

[ BTW to make the audio louder use Audacity’s (soft) limiter with make-up gain ]

Note that as written above, only the final notch filter is applied.

Also, the “S” character is now deprecated. To use the “S” variable to represent the selection in the current Audacity 2.1.2, you must enable the “Use legacy (version 3) syntax” option.

@Trebor: Did you intend to duplicate (notch2 s 420 15) and (notch2 s 914 15) or is that a typo?

To use the new (version 4) syntax, replace each occurrence of the “S” variable with “track”. For example:

 (notch2 *track* 942 5)

To apply all of the above filters, you can either:

  1. Apply one at a time. Example,
    Run
(notch2 *track* 285 5)

the run

(notch2 *track* 408 10)

and so on.

or
2) Nest the filters:

(notch2
  (notch2
    (notch2
      (notch2
        (notch2
          (notch2
            (notch2
              (notch2 *track* 942 5)
              914 15)
            914 15)
          420 15)
        420 15)
      408 10)
    285 5)
  204 10)

or
3) Write the code to re-assign the sound variable with the result of each filter:

(setf *track* (notch2 *track* 204 10))
(setf *track* (notch2 *track* 285 5))
(setf *track* (notch2 *track* 408 10))
(setf *track* (notch2 *track* 420 15))
(setf *track* (notch2 *track* 420 15))
(setf *track* (notch2 *track* 914 15))
(setf *track* (notch2 *track* 914 15))
(notch2 *track* 942 5)

or loop through a list:

(setf notch-list
  (list '(204 10) '(285 5) '(408 10)
        '(420 15) '(914 15) '(942 5)))

(dolist (nl notch-list *track*)
  (setf *track* (notch2 *track* (first nl)(second nl))))

The notch filters did each work when I tried them individually. I thought they would be applied successively if in a list.
But you’re right : only the last one on my list is being applied.
I’m sure I’ve used a list of notch filters in the past, and all were applied, [ I did tick the box for the “Legacy” syntax ].

New & improved notch list …

(setf notch-list
  (list '(102 5) '(203 15) '(285 15) '(420 15) '(920 15)
 '(942 5) '(994 15) '(1828 15) '(1880 15) '(2433 50)))

(dolist (nl notch-list *track*)
  (setf *track* (notch2 *track* (first nl)(second nl))))

new & improved notch list.png

Thank you very much, Trebor and Steve, for all the time and thought you’ve put into this for me. I’m working on it!