Permissions-Policy: on-device-speech-recognition directive

Experimental: This is an experimental technology
Check the Browser compatibility table carefully before using this in production.

The HTTP Permissions-Policy header on-device-speech-recognition directive controls access to the on-device speech recognition functionality of the Web Speech API.

Specifically, where a defined policy blocks usage, any attempts to call the API's SpeechRecognition.available() or SpeechRecognition.install() methods will fail.

Syntax

http

Permissions-Policy: on-device-speech-recognition=<allowlist>;

<allowlist>: A list of origins for which permission is granted to use the feature. See Permissions-Policy > Syntax for more details.

The default allowlist for on-device-speech-recognition is self.

Specification
Web Speech API # dom-speechrecognition-available

	desktop					mobile
	Chrome	Edge	Firefox	Opera	Safari	Chrome Android	Firefox for Android	Opera Android	Safari on iOS	Samsung Internet	WebView Android	WebView on iOS
`on-device-speech-recognition` Experimental	Chrome – Full support Chrome 142 (Release date: 2025-10-28) footnote Full support	Edge – Full support Edge 142 (Release date: 2025-10-31) footnote Full support	Firefox – No support Firefox footnote No support	Opera – Full support Opera 126 (Release date: 2026-01-08) footnote Full support	Safari – No support Safari footnote No support	Chrome Android – No support Chrome Android footnote No support	Firefox for Android – No support Firefox for Android footnote No support	Opera Android – No support Opera Android footnote No support	Safari on iOS – No support Safari on iOS footnote No support	Samsung Internet – No support Samsung Internet footnote No support	WebView Android – No support WebView Android footnote No support	WebView on iOS – No support WebView on iOS footnote No support

Tip: you can click/tap on a cell for more information.