Sliding Time Scale Help

Hi, I’m a relatively recent user and am struggling with an effect I’d like to apply (Audacity 2.0.5).

I’m working on creating several ‘werewolf’ howls for a theatre show. The source is a recording of a human actor pitch shifted down with a bit of reverb and is in the territory required. The human voice however just isn’t capable of a wide enough frequency range in a single breath.

What I’ve been trying to do is to use the Sliding Time Scale/Pitch Shift effect to raise the frequency toward the middle of the howl and then return back to the lower frequency from there to inject more range (and ‘chill factor’). I can’t get the middle point to marry up, if I select halfway, apply the effect (up) and then invert the selection and reverse the effect I end up with the gap pictured below. I don’t know what this represents and how (or if) it can be fixed. Anyone?

The obvious solution might be to split the parts and crossfade but the nature of this sound just leads to a weird sort of harmonic through the cross-fade which makes the edit obvious and which I also can’t fix.

Am now under a bit of time pressure so it would really save my bacon if somebody with more knowledge could suggest a path forward.

I’ve uploaded an example of my pitch-shifted source to just in case anybody might be inclined to play.

Thanks for taking a look here.

Does it sound OK irrespective of the “gap”? You are only seeing the gap (which is a split line) because you are zoomed into the maximum possible level. If you don’t like seeing the gap, merge the two clips by clicking on the black vertical line on the right edge of the “gap”, or select over the gap then Edit > Clip Boundaries > Join.


Hi Gale, thank you for pointing out how to join the tracks, I figured there had to be an option in some menu or other.

Unfortunately the point pops whether joined or not so I guess that’s my current challenge; how to apply the ascending and then descending pitch shift without ending up with an audible click on the point where I’m reversing the pitch shift (having taken care to ensure the shifts end/start with the same pitch change on both sides of the point).

I should say I did also try with a Time Track envelope but that then alters the speed so the peak is over far too quickly.

Thanks again.

I don’t think it’s easy to do if you are trying to apply the second time scale from higher pitch to original anywhere over the area from 2.5 s to 2.6 s. Really Sliding Time Scale wants an envelope or graph.

Although not what you want, I got a believable result by selecting 2.7 s to 3.25 s, Edit > Find Zero Crossings, apply one Time Scale from zero Initial to Final 4 semitones, zoom in on a minor click at the end of the selection and Effect > Repair the discontinuity, then use Envelope Tool to make 3.3 s to 3.45 s quieter.


The obvious solution might be to split the parts and crossfade but the nature of this sound just leads to a weird sort of harmonic through the cross-fade which makes the edit obvious and which I also can’t fix.

A short crossfade (a few or several milliseconds) usually takes care of this kind of thing.

Or, sometimes you can just make make sure the cuts/splices take place at a zero-crossing (you’d have to delete a few samples).

I’ve never attempted what you are doing, so I have no idea if this will work… But, maybe there’s a way to shift the pitch of the whole thing so the pitch in the middle is correct. Then, slide the pitch down at both ends without “touching” the pitch again in the middle? …I don’t know if you’ll still get a gap where the effect starts/stops.

Try using a “Time Track” to bend the pitch rather than a pitch shifting effect.
You will need to experiment to get the original howl the right duration because using a Time Track will change both the pitch and speed.
You may need to use a combination of the Time Track and pitch/tempo shifting to get the best effect.

I’d also suggest making a duplicate track (select the track, then Ctrl+D) and applying reverb only to the duplicate and applying it as “wet only” (only reverb and no “dry” sound). You can then use the envelope tool to shape the reverb so that there is a bit more reverb at the end of the sound than at the beginning.

Hello, just wanted to quickly thank you all for your thoughts on the problem, honestly I hadn’t even considered looking to make the transition on a zero crossing (rookie mistake). I hope to try everything suggested but may not actually have to as right now an acquaintance who has AudioSculpt (which can envelope pitch changes) is having a play for me first. Cheers!