Which voices are available?
Whatever is installed on your device. macOS and iOS supply Apple's system voices, Windows supplies the Microsoft ones, Android supplies Google's, and Chrome adds a set of network-backed Google voices on top. This page enumerates them with the Web Speech API and does not ship any of its own, which is why your list will differ from anyone else's.
Why does nothing play?
Three usual causes. The browser has no speech engine, in which case the tool tells you the API is unsupported. Or the voice list has not loaded yet, since it arrives asynchronously. Or your operating system has no voices installed for the selected language, which is common on stripped-down Linux installs and on some Android builds.
Can I download the speech as an MP3?
No. The Web Speech API renders audio directly to your output device and offers no hook to capture it, so no browser-based tool built on it can save a file. If you need an audio file you need a server-side TTS service, or you need to record your system audio with a separate application.
Is my text sent to a server?
This page uploads nothing. The caveat is that some of the voices your browser offers, specifically Chrome's Google voices, are synthesized remotely, so choosing one means Chrome sends the text to Google. Pick a voice your operating system provides (Apple on macOS, Microsoft on Windows) and the whole thing stays on the device.
How do I make it sound less robotic?
Change the voice, not the pitch. Each voice is a different synthesis model and the gap in quality between them is far larger than anything the rate and pitch sliders can do. Modern neural voices (the Siri voices on Apple platforms, the Natural voices on Windows 11) sound dramatically better than the legacy ones sitting next to them in the same list.