Which ML model does the WebSpeech API use?

Question

https://developer.mozilla.org/en-US/docs/Web/API/Web_Speech_API

1. Which ML model does the MDN WebSpeechAPI use?
2. Where is this model hosted?
3. Any limits on the number of free API calls? Any cost aspects to consider?
4. Is it open source?

Requesting insights on the questions above.

Thanks in advance!