Introduction to the Google Assistant Service

  • The Google Assistant Service provides a streaming endpoint for integrating the Assistant into your devices for experimental and non-commercial use.

  • You need a device with internet, microphone, and speaker to get started with the Google Assistant Service integration.

  • The guide provides step-by-step instructions for embedding and extending the Google Assistant using Python on a Raspberry Pi.

  • You can find resources for integrating with other languages and platforms using the provided integration guide.

  • The Google Assistant SDK cannot be used for commercial devices.

The Google Assistant Service gives you full control over the integration with the Assistant by providing a streaming endpoint. Stream a user audio query to this endpoint to receive a Google Assistant audio response.

You can generate bindings for this API from a variety of languages (for example, Go, Node.js, C++, and Java) on all platforms supported by gRPC.

Basic hardware requirements

Before you begin, you'll need the following components:

  • A device running on one of the supported platforms, with internet connectivity
  • A microphone
  • A speaker

Steps

The following sections are applicable to Python and the Raspberry Pi.

If you do not want to use Python, use the integration guide to integrate the Google Assistant Service in other languages and on different platforms.

Embed the Google Assistant

This section gets the Google Assistant Service sample working on your device:

  1. Set Up Hardware and Network Access
  2. Configure and Test the Audio
  3. Configure a Developer Project and Account Settings
  4. Register the Device Model
  5. Install the SDK and Sample Code
  6. Run the Sample Code
  7. Next Steps

Extend the Google Assistant

This section extends the Google Assistant Service sample to include Device Actions:

  1. Install Hardware
  2. Register Traits
  3. Handle Commands
  4. Add More Traits and Handlers
  5. Register Custom Device Actions