This library has been tested on OS X and Windows 10 in Unity 2019.1.4f1. At this time, I don't have the time and resources to add support for Android, iOS, or other platforms, but I welcome PRs that address this.
This library is available as a Unity package hosted on UPM. The easiest way to install it is via the UPM command-line tool (requires Node.js version 12 or higher):
# Install openupm-cli
npm install -g openupm-cli
# Go to your Unity project directory
cd YOUR_UNITY_PROJECT_DIR
# Install the package
openupm add com.oshoham.unity-google-cloud-streaming-speech-to-text
- In the Project Settings menu, change Player -> Configuration -> API Compatibility Level to .NET 4.x.
- Follow step 1 of Google's Cloud Speech-to-Text Quickstart Guide to:
- Set up a GCP Console project.
- Enable the Speech-to-Text API for your project.
- Create a service account.
- Download your service account's private key as a JSON file.
- Rename your private key JSON file to
gcp_credentials.json
. - Place your
gcp_credentials.json
file in a folder calledAssets/StreamingAssets
in your Unity project.
In your Unity scene, create a new GameObject
and attach the StreamingRecognizer
MonoBehavior to it.
If you want to quickly test that things are working, check the Enable Debug Logging
option on the StreamingRecognizer
, then play your scene. You should see some debugging messages appear in the Console, along with a live transcription of any speech audible to your computer's microphone.