Speakable (Beta) Schema

Speakable (Beta) Schema featured image
12 Oct 2020
Nirlep Patel
Structured Data

Currently, in the BETA version, the Speakable schema markup identifies various sections within an article or webpage that are best suited for audio playback using Text To Speech (TTS). With the help of the speakable schema markup search engines and other applications can easily identify reading aloud content over Google Assistant-enabled devices using TTS.

It allows web pages to distribute content through new channels and reach a wider base of users.Google Assistant uses the feature for answering news queries on smart speaker devices.

Whenever a user Google’s for news about a specific topic, the Google Assistant gives out three articles from the web and also supports the audio playbacks for various sections of the article using TTS with Speakable structured data.

PRO TIP 👨🏻‍🏫: Whenever Google assistant reads a speakable section aloud it credits the source and sends the full URL of the article to the user’s mobile device via Google’s assistant app.

Country And Language Availability:

Currently, The Speakable section is only available in the US for users having Google Home Devices set to English and publishers publishing content in English. Google hopes to launch the feature in more countries and languages very soon as soon as a sufficient amount of publishers implement the speakable entity.


  • Ensure to follow all the said guidelines.
  • Ensure to add the speakable structured data to your web page.
  • Submit your content for beginning the onboarding process. It can take up to 5 business days for reviewing your content. Google sends you a notification to know if you are eligible or not for the feature.


Apart from the Structured Data and Webmaster Guidelines, the following guidelines have to be followed:

👩‍💻Technical Guidelines

  • Avoid adding the Speakable structured data for contents sounding confusing in voice-only and voice-forward situations. Example: Photo Captions, Source attributions, etc.
  • When adding Speakable structured data, focus on key points as it allows listeners to have a clue about the story and not have the TSS readout eat out the important details.

✍️ Content Guidelines:

  • Content represented by Speakable structured data must incorporate precise headlines or summaries providing users with comprehensible and useful information.
  • In case you incorporate the top story in the speakable structure, it is recommended that you rewrite the top story to break up information into single sentences making it clear to read for TSS.
  • It is recommended to incorporate around 20-30 seconds of content per section of a speakable structure or around two to three sentences form optimal audio user experience.


Article or Webpage object use the Speakable Structured data.

  1. cssSelector: the entity holds addresses in the footnote.
  2. Xpath: Addresses content that uses Xpath ( it is assumed that the content has an XML view). You can use cssSelector or xPath, do not use both.


I cannot Trigger My Content, You cannot trigger your content via the Google Assistant using TTS audio, How do I fix the issue?

  • Try and use the below-listed commands:
  • What is the latest news about the said topic?
  • What is the latest news on the said topic?
  • Play News about the said topic.

If you are still facing trouble then it may be because the ranking is determined algorithmically. The Google Assistant will provide up to three articles belonging from different publications via TSS audio playback.

In case you are experiencing an issue not covered by the documentation

How do I fix the issue?

Make sure to visit the Google Webmaster Central Help Forum.

Make sure to contact your Google representative or audio news support.

Other Audio Solutions:

You can also use other Google Assistant Audio Solutions for your news content. Example: Advanced Integration of Google Assistant for your custom applications. Check out the Actions on Google Developer Guide.

PRO TIP👨🏻‍🏫 : To know in detail about fixing problems related to troubleshooting please refer to the previous posts of the series.