Bootleg with AI...

[Ancar]

Do you think we could play a little with the volume of Lou's voice in the Gymnasium bootleg? Strangely, I consider it too high. It overwhelms the mood of the band in some moments

[ybosde2]

Do you think we could play a little with the volume of Lou's voice in the Gymnasium bootleg? Strangely, I consider it too high. It overwhelms the mood of the band in some moments

I had a look to see if I could improve the vocal levels and overall sound of the Gymnasium tracks with mixed results (pun intended).

With GIFIL I gave a light boost to the vocals and used EQ to give them a more balanced sound. I also tried to even-out some of the sudden volume changes that occur in the recording and particularly in the vocal levels.

The results weren't too bad to my tin-ear.

Here's the result of my tampering:
https://drive.google.com/file/d/14HHnoW ... sp=sharing

And a clip from the original for comparison:
https://drive.google.com/file/d/1pvTiT4 ... sp=sharing

I then looked at "I'm Waiting for the Man" where the issue with Lou's vocals sometimes overwhelming the sound seemed more marked. For this I tried dropping the overall level of the vocals and using EQ to make them less tinny/harsh. The result was a better balance between vocals and the instruments, but a muddier overall sound. I preferred the original.

Perhaps someone who knows what they're doing can take this further, if there's any interest.

[ybosde2]

If AI could be used to create a vocal-less Live MCMXCIII, that would be a Very Great Thing.

This is what UVR5 does with it OOTB:
https://drive.google.com/drive/folders/ ... sp=sharing

Listening to just the vocal stems makes for an interesting experience.

[Ancar]

Great Job!

[Sheila Klein]

Has anyone tried applying reverb to the Hilltop Festival set? With no walls to echo off the recording sounds a little hollow to me, and might benefit from some artificial "room sound."