Speakable Schema

Speakable Schema is a Schema.org markup that marks text sections for search engines and voice assistants as suitable for reading aloud.

What is Speakable Schema?

Speakable Schema is a Schema.org annotation that marks certain text sections of a web page as especially suitable for text-to-speech (TTS). Voice assistants like Google Assistant can read these marked sections aloud when users ask for information by voice. The markup identifies the most concise and informative passages on a page.

Implementation occurs as JSON-LD in the head of the page with the type “WebPage” and the property “speakable.” You can use either CSS selectors or XPath to define the readable sections. Google recommends marking a maximum of two to three short sections per page — ideally concise summaries or definitions. The marked text should be understandable on its own, without needing to read the surrounding context.

For SEO, Speakable Schema is particularly relevant in the context of Conversational Search and AI Overviews. As more and more search queries are made by voice and AI systems cite text passages, a clear Speakable annotation increases the chance of being selected as a source for read-aloud or cited responses. Combine Speakable with additional Schema Markup like FAQPage or HowTo to make your content maximally machine-readable.