Treat this like phone design, not screen design. The user is holding an iPhone Pro Max one-handed.

Most thumbs only reach a certain range. Grade higher for anything reachable inside that thumb zone, and keep the primary actions near the bottom. The top 65 percent holds no important controls. It only shows what helps the user trust the system. The detected input, the transcript, the peripheral stuff they need to see but never have to touch.

Keep every interaction to about one and a half to two seconds. As simple as possible. A baby should intuitively understand what to do.

I am not asking you to design. I am telling you what is going on with my phone. There is a thumb. There is about a second and a half. A baby should figure it out.