As someone who's been doing AI call centres for a while, I'm not surprised. It's really, really hard to build voice AI that works. There are no open source solutions that work out of the box. But the biggest problem is ASR. WER is still atrocious even with SOTA models. When you add drug names and regional accents, it's a recipe for disaster.
Just curious, why are drug names such a disaster? Do LLMs not understand pronunciation keys?
Noooo come on We're not saying "Gloom" that's just the chemical.
Glumiphosphoroxate It treats mild whimsies.
Have one??????
YOU GOT ONE?
Come on in. Oh yeah. You have it. You definitely fucking have it LOL. Here take 2 of these.
Glumaway
You're one of us now.