1. People talk about an app call Yonomi, which I do not use, because I don't want too many 3rd parties having access to my smart home.
2. Then there's the Samsung Smart home integration.
3. If you have a Lutron Caseta hub already, you can easily create scenes for every function of Sonos which I have done. And then control those thru IFTTT and use specific phrases like, "Hey Google, Play Death Cab, Kitchen" short, simple and not adding a bunch of crud. Then I got to thinking, I don't want to use IFTTT, because I used this in the early stages IFTTT and results were finicky, but now they are pretty solid.
4. Now if you're a programmer with a raspberry pi, you can easily get the node-sonos api on GitHub and basically have a awesome set of features that have existed and people have asked for a long time ago. I'm pretty sure Yonomi borrowed from this source. You could send a broadcast long before the other smart speakers had this feature. You can have it speak a different language even, (Korean, Japanese, French...etc...). Anyway, I used this method because I want to get rid of IFTTT out of the equation when it was finicky. Of course, this method isn't going to be the easiest as I had to get a CA Authority Certificates and integrate Google's API.AI. But, I can control it from anywhere with internet access.