hard panned left & right vocal needs rectifying

It’s an old stereo mix containing one particular verse where - for a reason only known to the experimental me of the time - I hard panned the vocals first to the left for two lines and then to the right for two lines.

I don’t have the Master to remix it.

It is a mono female voice singing over a rather neat acoustic guitar pick which is in stereo centred.

Now when I try to manually pan centre the vocals it screws the nice sound of the guitar, likewise switching to mono for that segment.

The best I’ve got so far is using steve’s mixer plugin with the ‘narrow stereo’
preset, but I’d still like to get the vocals more central without losing too much from the balance of the guitar

Robert J. H. 's stereo-tool plugin has the option to isolate or remove the centre …
so may be of use.

Thanks Trebor, I tried all the presets and messing around a bit but it alters the tone of the acoustic guitar and makes it sound rather toytown as if it had been squashed and then passed through a chorus pedal.

I’d consider using it on an electric guitar or synth as an additive effect.

You can’t take apart a mixed performance into individual voices and instruments. That’s basically what’s required to remix a show into a more pleasant form. There are tricks you can play with stereo imaging and Left-Right directional clues, but those, as you’re finding are frequently not theatrically pleasing.

One thing you can do is reduce the show to mono, see if it sounds OK, and then build in “fake stereo” by splitting the mono into two matching tracks and then delaying one very slightly. That’s the ‘watching a live performance from the back of the room’ effect.

It may sound better than either what you have or the stereo imaging software tricks.

And yes, even if you have full theatrical, broad, Multi-Channel, Dolby Surround, etc, etc, the lead performers are almost always in the middle or close to it.


The distortion is rather the consequence of the stereo guitar than of anything that my tool does.
This is supported by the fact that the downmix to mono sounds strange to begin with.
There are a lot of tweaking possibilities and pre/post processing steps you can take.
A sample sound would really be helpful.

Hi Robert, thanks very much for the offer, but no can do for a sample. The vocalist doesn’t want it posted. I’m sure you could do a much better job but I still think it would come out sounding ‘processed’ with the sound quality on this segment differing from the rest of the song.

The setting ‘remove centre - no filter’ is the best for the guitar but then the vocals sound odd, a bit reverby, even though apparently coming over as more centred out loud.

But this is a bit of an illusion because on headphones faint reverb like echoes of the vocals can be heard on the opposite channel. The original is pretty dry.

I wouldn’t want to process the rest of the song as well to achieve continuity with the treated segment because I like it as it is.

This is what I’ve manually tried today:

1 - Split the stereo track
2 - Duplicate the channel segments containing the hard panned vocals
3 - Swap their channels to opposite side ie. left to right & right to left.
4 - Eq out as much of the low end of the guitar as possible on those segments.
(This also means sacrificing the vocal low end).
5 - Reduce the gain considerably on the copied segments to blend in subtly.

This doesn’t centre the vocals but does bring them up toward centre a bit more losing that hard left and rightness. The guitar that is also copied and swapped over along with the vocals is now very tinkly when heard in isolation with the vocal track it is on and neglible in its effect on the overall mix.

Ideally I’d like to centre the vocals a bit more but then the guitar mixed in with them would be louder and cancel out its stereo counterpart on the opposite track (if you are still following).

But this is already better than when the vocals were hard panned and hasn’t changed the nature of the overall sound save to fill the vocals out a little bit.

Thanks, doing that is not too bad a result but It lacks a certain something compared to the section it follows. It also ended up accenting the tape hiss dramatically. This song was mixed down from a Portastudio master onto a chrome cassette in 1988.

The hiss isn’t particularly noticeable when it is in stereo?

I would have to see if I have the first digital transfer version with me that includes some tape hiss lead-in to get a sample to use for removal. This wav copy doesn’t have any.

Ordinarily I don’t mind a bit of tape hiss so only get rid of it for lead-ins out lead-outs