This is crazy it does not do this.

Eight years. EIGHT years. EIGHT _YEARS_! Just precisely how far do you have your head up your backside that you haven't even assigned an intern code-monkey to this at some point in the last eight flipping years? Do you understand, at _ANY_ level, how stupid this makes your team look? In pretty much ANY app, for ANY reason, when the user makes a preference change, there is some simple way to immediately experience that preference change. Would you leave "language selection" or "dark mode" settings without some way to experience them BEFORE getting into a fully routed drive? NO. You let the user experience their choice immediately or almost immediately. Is your code base such garbage that it is actually DIFFICULT to get a selected voice to say something? I mean, it should just be adding "say('Hello, it's nice to meet you')". But how could it possibly take you -- and I know I'm getting extremely repetitive, but I have to say this again -- EIGHT DAMNED YEARS to do something that simple? Which suggests to me, a software engineer, that your code base is so hopelessly baroque and fragile that I shouldn't be trusting it with any decision making...

This has been a request since 2017. Just sayin'
https://www.waze.com/forum/viewtopic.php?p=1590038#p1590038

Why/How can this not already be a thing?

This can be enabled by allowing the change when the phone/vehicle is not in motion.

This! Please Waze!

This is an essential and basic feature!

I agree with the other commenters that it's a safety issue and makes ME (and others too) want to change voices once we're moving. I like changing voices, but often find that I can't hear or understand or the voice is just annoying and I HAVE to find another ASAP. Even just being able to hear (as a preview) the line they say as I'm starting the drive when I hit the name would be soooo very helpful and shouldn't take that much programing. Being able to hear a preview and not having to change after starting would prevent a lot of #mishaguyas

This is a basic safety issue. If we can’t hear a sample of the voice beforehand, it just encourages users to adjust their settings *while driving* around. Nobody needs that kind of distraction.

I have some hearing issues and I need to test voices to know if I will be able to hear them.