What the browser can say aloud, in its own voices, through the Web Speech API, which needs no permission. Its voices read text as it is spelt: none reads IPA, nor the SSML that would carry it. A page, an extension's pages and its content scripts have them, but not a worker, as an extension's service worker is. It reads the global speechSynthesis as it is asked.
Says text aloud in a voice, cutting short whatever the browser was saying, even what a page itself was, as it says one thing at a time. It begins in its first step, so that it still has the user's activation, which a browser may need to speak, when run at once from an event's handler (as IO's syncStep runs it). Cancelled, it stops saying it.
Says text aloud in a voice, cutting short whatever the browser was saying, even what a page itself was, as it says one thing at a time. It begins in its first step, so that it still has the user's activation, which a browser may need to speak, when run at once from an event's handler (as IO's syncStep runs it). Cancelled, it stops saying it.
Nothing, once the text has been said, or cut short by something else said. It fails with a SpeechError where the text cannot be said: the browser has no speech, no longer has the voice, or the voice fails.
The browser's voices, once it has listed them: at once if it has, or else once it does, as Chromium lists none until first asked, waiting at most Speech.patience.
The browser's voices, once it has listed them: at once if it has, or else once it does, as Chromium lists none until first asked, waiting at most Speech.patience.
Attributes
Returns
The voices, in the browser's order, none where it has no speech.