How to add text-to-speech to WordPress

Decide whether visitors should hear content through their own browser or whether your site should serve generated audio. Browser speech is quick and can avoid an external service, but voices, language support and pronunciation depend on the visitor’s device and browser. The Web Speech API exposes speech synthesis through JavaScript, including voice selection, rate and pitch controls.

Use generated audio when you need consistent playback, downloadable files, multilingual narration, analytics or audio that remains available after the page has loaded. A browser-only approach does not create an MP3 for visitors to download.

Decide exactly what should be read. Use the article’s main content container rather than the whole page, otherwise the player may read navigation, cookie notices, comments or related-post links. Give the content area a stable class or selector that your script or plugin can target.

Keep a normal text version on the page. Text-to-speech is an additional way to access content, not a replacement for headings, readable HTML, keyboard access or screen-reader support.

For a no-service implementation, add a visible button and JavaScript through a child theme or a small custom plugin. The script should collect the text from the article container, create a SpeechSynthesisUtterance, and send it to speechSynthesis.speak(). Add separate controls for pause, resume and stop, using the corresponding speech-synthesis methods.

Load available voices with speechSynthesis.getVoices(), then refresh the voice list when the browser fires voiceschanged. Provide sensible defaults for language, rate and pitch, and handle browsers that return no voices or do not expose speech synthesis.

Test the browser version in Chrome, Edge, Safari and Firefox on both desktop and mobile. The API is widely available, but the actual voice list and voice quality vary by browser, operating system and installed language packs. Some browsers also populate voices asynchronously, so a voice dropdown can be empty until the voices list changes.

Do not put script tags directly into a post or page. WordPress may remove them, and allowing arbitrary script in content creates a security risk. The manual approach also leaves you responsible for selectors, player styling, keyboard focus, content changes and troubleshooting conflicts with caching or page-builder markup.

If you need a maintainable site-wide setup, install a text-to-speech plugin from Plugins in the WordPress dashboard, activate it, and choose which post types should receive a player. Configure the default language, voice, reading speed, pitch and player position, then use a block or shortcode when you want the player in a specific location. Many plugins can automatically add players to posts and pages, while shortcode-based tools give you more control over placement.

If the plugin generates audio through a third-party service, add the required API key in its settings rather than in an article or page. Check whether audio is generated when you publish, on demand, or only after you manually start a conversion. Keep an eye on usage limits and privacy disclosures.

Open a published post as a visitor and test play, pause, resume and stop. Check that the player reads only the intended article, that headings and lists sound in the right order, and that the button has a clear label and visible keyboard focus.

Test a short post, a long post, a multilingual post and a post containing links, quotes, tables or page-builder elements. If generated audio is stale after editing an article, regenerate it or enable the plugin’s update-on-change option where available. If the main requirement is multilingual voices, adjustable speed and pitch, an embeddable player and engagement tracking, Speaker is a suitable plugin route for this job.