SpeechRecognition: install() static method

The install() static method of the SpeechRecognition interface installs the language packs needed for on-device speech recognition of the specified languages, at the specified quality level. A language pack is the data a browser needs to recognize speech in a particular language. Once it is installed, recognition in that language can run on the user's device, so audio doesn't have to be sent to a remote service.

To check whether the language packs are already installed, use the SpeechRecognition.available() method with processLocally set to true.

Access to the install() method is controlled by the on-device-speech-recognition Permissions-Policy directive. If a policy blocks access, the promise returned by install() rejects with a NotAllowedError DOMException.

Syntax

js
install(options)

Parameters

options

An object specifying options for the installation. Possible properties include:

langs

An array of one or more strings containing BCP 47 language tags, each representing a language that you want to install the language pack for.

quality Optional

An enumerated value that indicates the approximate use case of your speech recognition app. install() installs language packs that support at least this quality level. Possible values are:

command

Level 1: Short isolated phrases with limited vocabulary and a single speaker. Use cases include voice commands for apps. This is the default value.

dictation

Level 2: Continuous speech with moderate background noise and a single primary speaker. Use cases include dictating long-form text inputs such as SMS messages, email bodies, or strings for translation.

conversation

Level 3: Continuous speech with complex vocabulary, high background noise, and multiple primary speakers. Use cases include transcribing meetings and continuous live captioning.

processLocally Optional

A boolean. install() always installs language packs for on-device speech recognition, so this option has no effect. This means you can pass install() the same options object that you pass to available(). See Browser compatibility for browser-specific differences.

Return value

A Promise that resolves with a boolean value indicating whether the language packs were installed successfully. The promise resolves with:

true

All installation attempts succeeded for the requested languages, or the languages were already installed.

false

One or more of the requested languages aren't supported, the language pack for a supported language failed to install, the browser declined to download a language pack (for example, because the user refused a permission prompt), or the langs array was empty.

Exceptions

InvalidStateError DOMException

The current document is not fully active.

NotAllowedError DOMException

One of the following:

SyntaxError DOMException

One or more of the strings specified in langs aren't valid BCP 47 language tags.

Examples

Checking on-device availability and installing language packs

For on-device speech recognition to work, the browser must have a language pack installed for the language you want to recognize. If you call start() after setting processLocally to true but the correct language pack isn't installed, recognition fails and an error event fires with a service-not-allowed error code.

To install the correct language pack, follow these two steps:

  1. Use the SpeechRecognition.available() method to check whether the language pack is installed on the user's device.
  2. If it isn't installed, use the install() method to install it.

These steps are handled using the following code snippet:

js
startBtn.addEventListener("click", () => {
  // Check availability of target language
  SpeechRecognition.available({ langs: ["en-US"], processLocally: true }).then(
    (result) => {
      if (result === "unavailable") {
        diagnostic.textContent = `en-US not available to download at this time. Sorry!`;
      } else if (result === "available") {
        recognition.start();
        console.log("Ready to receive a color command.");
      } else {
        diagnostic.textContent = `en-US language pack downloading`;
        SpeechRecognition.install({
          langs: ["en-US"],
          processLocally: true,
        }).then((result) => {
          if (result) {
            diagnostic.textContent = `en-US language pack downloaded. Try again.`;
          } else {
            diagnostic.textContent = `en-US language pack failed to download. Try again later.`;
          }
        });
      }
    },
  );
});

We first run the available() method, specifying one language (langs: ["en-US"]) to check availability for, and processLocally: true. We test for three different possibilities of the return value:

  • If the resulting value is unavailable, it means that no suitable language pack is available to download. We also print an appropriate message to the output.
  • If the resulting value is available, it means that the language pack is available locally, so recognition can begin. In this case, we run start() and log a message to the console when the app is ready to receive speech.
  • If the value is something else (downloadable or downloading), we print a diagnostic message to inform the user that a language pack download is starting, then run the install() method to handle the download.

We then call the install() method with the same options. It starts downloading the en-US language pack and returns a Promise that resolves with true if the language packs were downloaded and installed successfully, or false if not.

This code is excerpted from our on-device speech color changer (run the demo live). See Using the Web Speech API for a full explanation.

Installing a language pack for a specific quality level

The following code snippet modifies the previous example. It calls available() with the quality option set to dictation, to check whether on-device recognition supports this quality level. If the result is unavailable, we set the SpeechRecognition object's processLocally property to false (assuming it was previously true), which allows the browser to use a remote recognition service if it has one, and then call start().

If the result is available, we call start() to begin on-device recognition. If the result is any other value, we call install() with the quality option set to dictation to install the required language packs.

js
startBtn.addEventListener("click", () => {
  // Check availability of on-device target language dictation quality
  SpeechRecognition.available({
    langs: ["en-US"],
    processLocally: true,
    quality: "dictation",
  }).then((result) => {
    if (result === "unavailable") {
      diagnostic.textContent = `On-device recognition for dictation not available. Falling back to any available recognition service.`;
      recognition.processLocally = false;
      recognition.start();
    } else if (result === "available") {
      recognition.start();
      console.log("Ready to receive a color command.");
    } else {
      diagnostic.textContent = `en-US language pack downloading`;
      SpeechRecognition.install({
        langs: ["en-US"],
        processLocally: true,
        quality: "dictation",
      }).then((result) => {
        if (result) {
          diagnostic.textContent = `en-US language pack downloaded. Try again.`;
        } else {
          diagnostic.textContent = `en-US language pack failed to download. Try again later.`;
        }
      });
    }
  });
});

Specifications

Specification
Web Speech API
# dom-speechrecognition-install

Browser compatibility

See also