See what AI says about your brand. Get your free
Speakable Schema is a structured data type that identifies specific sections of a webpage as particularly suitable for text-to-speech audio playback, designed originally for voice assistants and smart speakers. The markup signals to AI and voice systems which portions of a page are most appropriate for reading aloud, typically the key summary, main answer, or most important sections of the content. It helps AI voice systems avoid reading unnecessary navigation text, boilerplate, or secondary content and focus on the most substantive parts of the page.
Speakable Schema directly feeds AI voice search by giving voice-enabled AI systems a clear signal about which part of your content to read when responding to a spoken query. As LLM-powered voice assistants become more capable of drawing from web content, the speakable sections of well-marked-up pages are more likely to be selected and read aloud as the answer. This makes Speakable Schema increasingly relevant beyond its original context of Google Assistant integration, applying to any AI voice interface that retrieves and reads web content as part of its response generation.
Why it matters: Voice-based AI queries are growing as Siri, Alexa, and Google Assistant integrate more sophisticated LLM capabilities. Speakable Schema positions your most important content to be the answer that gets read aloud.