Voice Acting Ponderance

So for voice acting, there's two common approaches.

If you have the budget -- fully acted out lines. Maybe not the whole game because for a narrative game that'd be tens of thousands of dollars, but at least some. If not, visual novels use "beepers" to mimic the sound of someone talking, or at least represent it aurally in the same way a thumbnail represents a painting.

For The Shadow Over Cyberspace, I drew a line and said "Anything taking place in the real world is fully acted; anything in cyberspace is a beeper." Worked pretty well! But now we're moving on to Laugh Track, the next game, and I'm hitting an conundrum.

There isn't an easy line to draw to say "Everything here is acted, everything there is not" like with SOC. So I'm thinking of a compromise -- Simlish.

If you've played The Sims, you know what Simlish is. It's a sort of nonsense gibberish the characters speak, randomized but tuned to convey emotion even if it conveys no meaning. It's a bit like Animal Crossing's "animalese" in that regard, but more sophisticated. I have voice actors record random nonsense with various emotional tilts, and use that for the characters instead of beepers or full lines.

Here's the problem, though. In the past, I've aimed to cast representatively. If a character is African-American, get an AA voice actor. If a character is Hispanic, get a hispanic voice actor. And so on. But I also have streamers and members of the community eager to work with me on this, which reduces casting (and honestly costs) quite a bit... but would mean no representative casting.

So I pose this question to you: How important is representative casting if the "dialogue" is all gibberish anyway? How do you feel about this?

I'm genuinely curious. Leave comments below or reach out directly to me. I'd like to get a feel for what the community wants from this game, so I'm being practical while not violating goodwill. Thank you.