Trying to figure out what processing is being done on Kesha's voice here...

https://www.youtube.com/watch?v=edP0L6LQzZE

It's only on the main lyrics, not the chorus, and not all the time.. mostly in the first minute of the video.

Definitely some auto-tune, but Kesha has a fairly thin voice normally and there's something putting a really hard edge on it here. Sort of like the "Britney Spears Bandpass Filter" in reverse.

Wondering if it's maybe mixing auto-tune output with unprocessed voice to give a bit of a chorus/phasing effect... or maybe just a backup singer layered in on top

No idea but she's singing about my place.

Sounds like a standard 80 vocoding effect. i guess that what was once old is now new again

Hah, you're right... I missed the vocoder sound at the start so wasn't as precise as I intended.

I was talking more 0:15-0:30 and 1:24-1:40, sorry about that.

During the break at 0:45 it sounds like maybe shifting her voice down a couple of steps (or could just be another singer)... maybe during the leads they're mixing some of that frequency-shifted vocal in, but it doesn't quite sound like that. Think it's more likely to just be auto-tuned and non-auto-tuned vocals mixed together.

Or maybe I have successfully avoided auto-tuned vocals for long enough that I have forgotten what they sound like without any other effects.

Yes. I know. It is the same thing that Cher used with that life after love song. it's still a vocoder. At the very beginning they are vocoding to a set of chords so you get the familiar syfi computer voice that was clearly early eighties. but then you get into the 0:15~0:30 point and they are doing single note vocoding where you can hear the clear 2-3key pitch progression happening.

There's a hiphop group out of Winterpig called the Lytics. They like to play with autotune on their records, more comedy style. Funny voices.
Sadly most who hear anything related to such sounds automatically think of Life after love. When it was totally 80s she borrowed it from.

Interesting... thanks !

It makes sense that they would use the same effect in both places.