automated silence interval

Win 7 64 bit
Aud 2.05 exe

How do I edit audio files to alternate silences between the left and right channels at regular intervals like every 0.5, 0.6 or 0.7 seconds?

I want to experiment with many different intervals with many different files so I would prefer to be able to automate the editing process, if possible.

If anyone’s curious this is for applications of EMDR.

I’m not entirely sure how I would do that manually, much less automated.
Import or Open the show. Using the drop-down on the left, Split Stereo Track.
Select the top track and drag-select 1/2 second. Silence it with Control-L. Select the next 1/2 second on the bottom track and silence that with Control-L. Select the third segment on the top track and silence it. Repeat forever.


You could try the “Panning (LFO)” plug-in available here:
To install Nyquist plug-ins, see:

Close! Thanks.

I need squared sine waveforms or the ability to make my own. The closest I could find is this but its Linux:

Is anyone aware of any plugins that might help me?

You can do it with this Nyquist script in the “Nyquist Prompt” effect (Effect menu).

(setq frequency 0.5)
(setq bias 0.0) ; must be between -1 and +1. 0 for square.

(if (arrayp s)
  (let ((lfo (osc-pulse frequency bias)))
      (mult (aref s 1)(sum 0.5 (mult 0.5 lfo)))
      (mult (aref s 1)(sum 0.5 (mult -0.5 lfo)))))
  "Error. Stereo track required.")

Copy and paste that code into the Nyquist Prompt effect and apply to a stereo track.

If your track is currently (single channel) mono you will need to convert it to a 2 channel track first (duplicate the track (Ctrl+D) then join the two tracks together into a stereo track

I found a way to do it with Isochronic Modulator plugin:
snap to: seconds
pulse width 50%
mod freq 1.5
split stereo, shift one by a second

Will check out your solution steve, thx

works like a charm…thx

If you are going to do a lot of this, you could quite easily convert that code into a plug-in. Please ask if you are interested, or go here if you want to work that out yourself:

Although Steve has already posted a functional snippet, Here is my version.
You can set the hardness from smooth (sine tone) to hard (square).

(Defun Pulse (snd)
  (Mult snd (Sum 0.5 (Clip (Scale (+ HARDNESS 0.5) 
     (Osc (Hz-to-step (/ 0.5 DUR)) 1 *TABLE* PHASE)) 0.5)))) 
;;; Main
(Psetq PHASE 180; Fixed for Alternation 
       DUR 1.0; In Secconds
       HARDNESS 50); 0=sine, ~10000 = Square
(Multichan-expand #'Pulse S)

By the way, The bias from Steve’s example changes the duration ratio left to right.

That code fulfils the brief , but needs a “scale 0.5” , as otherwise the resultant waveforms double in amplitude.

That type of alternating code could be useful for before-after demonstrations : chopping between the original and the processed version , see attached …

Thank you Trebor, I’ve corrected the code in order to be on the safe side.
Normally, the two channels should each have a gain of -6 dB and the doubling - if the other channel is silenced - shouldn’t be a problem.
There are different strategies for proper panning thinkable.
A dual mono sound should be panned from 0 dB (all left) to -6 dB (all in the middle, overall energy still 0 dB) to 0 dB (all right).
If different sounds are allocated to the two channels, -3 dB will be the chosen value for center panning.
Furthermore, the distribution should actually be different for head phones or speakers. The latter should regard the angle between listener and speaker 1 and 2. This is normally 60 degrees.
Another interesting fact is that the extreme panned sound is equally located at the farther left or right if the gain difference is about 18 dB.
Our snippets silence the opposite channel totally whereas a scale factor of 0.1 would be enough (and probably better for the speakers). Many head phones show a comparable “leakage”.