Google added to its help docs an explanation on how to opt out of Google Read Aloud with a meta tag, although the meta tag is not new. Google also clarified in those help docs the crawling behavior of the user agent.
Again, the useargent for google-read-aloud is not new, it is from 2019 or older. But I think it is different from the speakable markup, because Google said in the help doc that the google-speakr agent is an older version of the user agent that is now deprecated. Its new name is Google-Read-Aloud, so I wonder what is up with Speakable markup, last we heard it is not dead yet and the speakable markup doc is still active.
In any event, Google added details on how to prevent Google Read Aloud from accessing your content saying “To prevent paywalled content from being read aloud, use structured data for subscription and paywalled content. Make sure the isAccessibleForFree property is set to False. To opt out of Google Read Aloud completely, use the nopagereadaloud meta tag:
Google also added a section for the crawl frequency and behavior of Google Read Aloud. Google said Google Read Aloud is triggered by a user request. Google Read Aloud conserves bandwidth by caching page results, but you may still see multiple requests for a given page. Google Read Aloud is not a web crawler: it is activated upon user request, and it doesn’t follow links. A user request for listening to a web page may or may not result in page visits, depending on whether the page has been fetched recently.
The meta tag for nopagereadaloud is old, has been around for a while – Google is just clarifying how to use it in its help documents for about a year.
Forum discussion at Twitter.