Download and Setup
When your Capture Studio build has been completed, it will be uploaded to Google Drive for you to download.
You will need to provide us with the email address of the account that will be given access to the Google Drive hosted downloads.
This is a pre-requisite before you can complete any more steps.
Once we have your email address, you will be provided access to three downloads.
- Download and unpack the build .zip file in a location of your choosing
- Download and unpack the two .tar.gz files into /DockerServices (contained within the unpacked .zip build folder structure)
The zip will include two folders:
- Docker Services
- Files for setting up the Docker services required to have the digital human speak
- Windows
- The Capture Studio application files
The following setup steps are only required if you wish to have the digital human speak. If you don't need the digital human to speak, you can skip all the setup steps.
The Capture Studio is an early access product and for this reason, the installation process is not yet fully streamlined. There may be issues and unexpected behaviour in the installation process. Please let us know if you encounter issues.
Docker is a third party system for running and managing applications.
We use Docker Desktop to make it easy to run all the ancillary services that are needed for the Capture Studio.
Steps:
- Follow the prompts in the installation wizard
- If prompted, selected WSL2 instead of Hyper-V. You won't always be prompted to make this selection
- Add your credentials to the /DockerServices/tts-service/.env file
- There are credential options for Azure TTS or ElevenLabs. Complete the credentials for the TTS provider that you wish to use. You will need to have registered with the TTS provider and procured API credentials from the provider which will be available via their admin interface
- Ensure Docker Desktop is running. It must be running for the next two steps.
- Run /DockerServices/LoadImages.bat
- Make sure the .tar.gz files are in the same folder as the LoadImages.bat
- This can take some time to run. Up to ~45 minutes and will not show any progress. Just leave it to complete the process, do not cancel the process
- Note: You can check if this step was successful by navigating to the Images tab in the Docker Desktop application and looking for the two ~30GB images
- Run /DockerServices/RunServices.bat
- Check in the Docker Desktop interface that all the containers are running. You should see a set of green containers showing they are running. See image below
Your voice selection is configured via an .ini config file. This Config file is set to Read Only because Unreal will overwrite it in certain situations.
Turn off Read Only when editing and then turn Read Only back on after editing. You can do this by Right Clicking the file in Windows Explorer and opening Properties.
- Open /Windows/KioskRenderer/Saved/Config/UneeQCore.ini
- Modify the fields as desired to choose your voice to use in the Capture Studio
Example settings in UneeQCore.ini:
Once the above steps have been completed, you're ready to run the Capture Studio.