There are actually a couple of things going on here, lot of disparate tricks so fair warning, it's gonna take a lot of practice to get that sound "competently".
You can definitely do similar things in Ableton, but I would advise either Autotune or Melodyne. The "graph" mode of either will give you a split between words and notes and allow you to drag them around, change the formant characteristic and related vocal timbre effects. Typically the pitching is done on each word or syllable, only occasionally multiple times in a word or syllable.
A key thing to this effect (and why Autotune or Melodyne excel vs a DAW) is that these two artists typically use a more slow pitch drift than an electro producer would (ie Skrillex). So it is important to fine tune the "drift" of the pitch moving to and away from your source and destination note. Autotune and Melodyne excel @ this. This makes it sound like a real vocal performance, but kind of like an alien one.
Another key thing to note is that there is actually a fair bit of harmony in play. Some of the notes that stand out, sound exceedingly pleasant or really dissonant are that way because they are really multiple copies of the word playing a chord (say a triad... root + 3rd + 5th) mixed really well to sound like one source. Also you can experiment with having the chord harmonics fade in and out of the root, or get a nice effect by arpeggiating a chord in the same word (think the "know" in "Tell me how to kno-o-ow" and then resolving on the second repetition. also the pad sound that intermittently plays that sounds like "ahhhhh" - probably multiple copies of the vox or the vox with a synth playing harmony).
I have a good friend who is amazing at all of this using just Ableton... but in the more skrillexy style. If you wanna do this style in Ableton you probably need to be real good with the pitch automation and volume envelopes, as each word needs to flow naturally here.