Glossary

Speakable Schema

Speakable schema is a type of structured data markup that identifies sections of a webpage most suitable for text-to-speech conversion by voice assistants and screen readers.

Get in touch Learn more

Data scientist building training data pipeline on laptop, data preprocessing visible, technical workspace.

TEXT-TO-SPEECH STRUCTURED DATA

What is Speakable Schema?

Speakable schema is a type of structured data markup that identifies specific sections of a webpage most suitable for text-to-speech conversion by voice assistants and screen readers.

Speakable schema is a SpeakableSpecification structured data type within Schema.org that explicitly marks content sections—such as headlines and summary paragraphs—as optimal for text-to-speech (TTS) playback. By implementing this markup in JSON-LD or microdata, publishers guide voice assistants like Google Assistant to read aloud only the most relevant, concise portions of a page, filtering out navigation, ads, and tangential content.

This specification uses cssSelector or xpath properties to pinpoint the exact DOM elements containing speakable content. For answer engine optimization, speakable schema is critical because it directly influences how AI-generated voice responses are constructed from web sources, ensuring that zero-click content delivered via smart speakers is coherent, authoritative, and free of extraneous noise.

STRUCTURED DATA FOR VOICE

Key Features of Speakable Schema

Speakable schema markup identifies sections of a webpage most suitable for text-to-speech conversion, enabling voice assistants to deliver clear, concise audio responses.

Core Specification

The speakable property, part of the Schema.org vocabulary, is applied to a WebPage or Article entity. It uses xpath or cssSelector to pinpoint specific HTML elements containing text optimized for TTS. This ensures assistants read only the most relevant summary, not navigation or ads.

EXPLORE

TTS-Optimized Content Selection

Marked content must be concise and conversational. Best practices include:

Targeting the article's lede or a dedicated abstract
Using short, declarative sentences
Avoiding ambiguous pronouns without clear antecedents
Omitting complex tables or data visualizations that cannot be read aloud meaningfully

Implementation Methods

Speakable can be implemented via JSON-LD, Microdata, or RDFa. JSON-LD is the recommended format by Google. A typical implementation specifies @type: SpeakableSpecification and includes a cssSelector array pointing to the target DOM elements, such as [".article-summary", "#tts-content"].

Platform Support & Limitations

Google Assistant is the primary consumer of speakable schema, using it to power News on Assistant and other audio briefings. The feature is currently limited to news content and requires publishers to be approved for the Google News platform. Other voice assistants have not yet widely adopted this specific markup.

Relationship to SSML

Speakable schema identifies what to read, while Speech Synthesis Markup Language (SSML) controls how it is read. SSML provides fine-grained control over pronunciation, pitch, rate, and pauses. For advanced voice experiences, speakable content can be paired with SSML to improve prosody and naturalness.

SEO & GEO Impact

While not a direct ranking factor, speakable schema enhances Generative Engine Optimization by providing a clean, authoritative signal for voice-based answer engines. It increases the likelihood of being selected as the single definitive source for audio responses, improving brand presence in screenless environments.

Enabling Efficiency, Speed & Accuracy

Intelligent Analysis, Decision & Execution

We build AI systems for teams that need search across company data, workflow automation across tools, or AI features inside products and internal software.

Talk to Us

Search across company data

Give teams answers from docs, tickets, runbooks, and product data with sources and permissions.

Useful when people spend too long searching or get different answers from different systems.

Enterprise searchRAGPermissions

Automate internal workflows

Use AI to route work, draft outputs, trigger actions, and keep approvals and logs in place.

Useful when repetitive work moves across multiple tools and teams.

AI agentsWorkflow automationGovernance

Add AI to products and internal tools

Build assistants, guided actions, or decision support into the software your team or customers already use.

Useful when AI needs to be part of the product, not a separate tool.

AI integrationDecision supportModel routing

SPEAKABLE SCHEMA

Frequently Asked Questions

Clear, concise answers to the most common technical questions about implementing Speakable structured data for voice assistants and text-to-speech engines.

Speakable Schema is a type of SpeakableSpecification structured data markup, defined by Schema.org, that identifies specific sections of a webpage most suitable for text-to-speech (TTS) conversion by voice assistants like Google Assistant. It works by wrapping content in application/ld+json blocks using @type: SpeakableSpecification and xpath or cssSelector properties to point to the exact HTML elements containing the text to be read aloud. When a voice assistant processes a page, it parses this markup to extract only the highlighted passages, ignoring navigation, ads, and other non-essential content, delivering a clean, concise audio summary to the user.

SPEAKABLE SCHEMA ECOSYSTEM

Related Terms

Speakable schema does not exist in isolation. It is a critical component within a broader ecosystem of voice search, structured data, and AI-driven content delivery. The following concepts are essential for understanding how speakable markup integrates with modern answer engine optimization.

Schema Markup

The foundational semantic vocabulary that Speakable extends. Schema.org markup uses tags added to HTML to help search engines and AI crawlers understand the entities, attributes, and relationships within content. Speakable is a specific property within the broader Schema.org hierarchy, relying on the same JSON-LD or Microdata syntax to define which textual passages are optimized for text-to-speech conversion. Without a robust schema strategy, speakable annotations lack the entity context needed for accurate voice responses.

EXPLORE

JSON-LD

The preferred serialization format for implementing Speakable schema. JSON-LD (JavaScript Object Notation for Linked Data) is a lightweight method to embed structured data within a <script> tag in the page head or body, making it easily parsable by Google and other voice assistant platforms. A speakable implementation using JSON-LD isolates the TTS-optimized content from the visual presentation layer, allowing developers to specify a cssSelector or xpath to the exact DOM elements containing the speakable text.

EXPLORE

Answer Engine Optimization (AEO)

The holistic practice of designing content to be the single, definitive source for AI-generated direct answers. Speakable schema is a tactical instrument within an AEO strategy, specifically targeting voice assistant and smart speaker interfaces. While general AEO focuses on featured snippets and generative overviews, speakable markup ensures that when a voice assistant reads an answer aloud, it selects the most natural-sounding, concise passage rather than a verbose or poorly formatted text block.

Conversational Search Adaptation

The discipline of optimizing content for natural language, multi-turn queries typical of chat-based AI interfaces and voice search. Speakable schema directly supports this by identifying content that mimics spoken language patterns. Key considerations include:

Prosody: Ensuring the marked text has natural cadence when read aloud
Conciseness: Voice answers must be shorter than text-based snippets
Context Independence: The speakable passage should make sense without visual cues like images or tables

Semantic HTML5

The use of modern HTML elements like <article>, <section>, and <aside> to convey explicit structural meaning. Speakable schema often uses CSS selectors that target these semantic elements to identify the main content body. Proper semantic HTML5 authoring ensures that the xpath or cssSelector defined in the speakable markup reliably points to the intended content, preventing voice assistants from reading navigation menus, footers, or advertisements as part of the answer.

Featured Snippet Optimization

The process of structuring web content to be selected for a prominent answer box at the top of organic results. While Featured Snippets serve text-based queries, Speakable schema serves voice-based queries. The two are converging: Google often sources its voice answers from the same passages that win featured snippets. Optimizing for both requires:

A clear, concise answer within the first 2-3 sentences
Structured data markup for both Speakable and FAQ or HowTo schema
Content that reads naturally when spoken aloud

About the author

Prasad Kumkar

CEO & MD, Inference Systems

Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.

His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.

Limited slotsGet a Free AI Consultation

How We Work

Custom AI workflows for your Business

One-fit-all AI don't work for modern businesses. At Inferensys, we aim to understand your business & custom requirements; which we use to define most efficient agentic workflows, the data, and the tools for your business.

Talk to Us

Speakable Schema

What is Speakable Schema?

Key Features of Speakable Schema

Core Specification

TTS-Optimized Content Selection

Implementation Methods

Platform Support & Limitations

Relationship to SSML

SEO & GEO Impact

Intelligent Analysis, Decision & Execution

Search across company data

Automate internal workflows

Add AI to products and internal tools

Frequently Asked Questions

Schema Markup

JSON-LD

Prasad Kumkar

Partnered with leading AI, data, and software stack.

Custom AI workflows for your Business

Review the use case

Pick the right approach

Build the first useful version

Improve from there