How does Closed Captioning work?

Here’s how the closed captions (CC) work right now:

Presenter (the one whose voice we want as captions - first tab):

The presenter selects the switch and specifies the language they are speaking.
The accuracy of word recognition depends on the browser and language.
The browser performs speech-to-text, and the captions are sent to the platform in the selected language.

Audience (participants - second tab):

For participants to see captions, the presenter must send them to the platform (as explained above), and the participant needs to enable the option to receive captions.
By default, captions appear in the same language the presenter sent to the platform. If the presenter speaks English, captions will be in English.
Participants can enable a translation feature in the second tab. The platform will then try to translate the captions before displaying them, using the same mechanism as the chat message translations.
If the speech-to-text language option isn’t visible in the first tab, it might not be supported by the browser.