Google Cloud’s Vision API allows you to use the Vertex AI Vision environment to build and manage applications. Vision also uses machine learning models to process and interpret images. In Glide, you can use the Google Cloud Vision integration to extract text from PDFs and image files.
Don't see the Google Cloud Vision integration?
You may need to upgrade your plan. Browse Glide's plans and find the right fit for you.
In Glide, click the Settings icon in the upper-right corner.
Navigate to the Integrations tab and then select Google Cloud Vision.
Click the Add button.
Enable the Vision API on your Google Cloud Project.
Navigate to your Google Cloud Console API Credentials.
Create and copy a new API Key for Glide.
Back in Glide, add your API Key in your Google Cloud Vision integration settings.
Features
The Google Cloud Vision integration in Glide currently supports two features:
Extract text from PDF files
Extract text from image
Extract text from PDF files
The Extract text from PDF files feature lets you select a PDF file, and Google Cloud Vision will extract text from the document. You can then use that text in your app. Google Cloud Vision has a five page limit on extracting text from PDF files, so only five pages can be extracted at a time.
The action can be added to a Component that supports Actions (e.g., a Button component), added to an action sequence in the Action Editor.
The Extract text from PDF feature is not available as a computed column.
If you have not yet set up a URL column to store the PDF file links, and a text column to store the extracted text, do that first in the Data Editor.
Select the component you'd like to add the action to. Remember, this must be a component that supports actions.
Search for the Extract text from PDF files action or navigate to Integrations -> Google Cloud Vision -> Extract text from PDF files.
Name the action.
Choose an icon.
Select the column with the File links.
Choose the language ID for the text. A full list of languages and IDs can be found here.
Choose the model, Document or Text.
Choose which pages to extract, or leave blank to extract all.
Specify where to store the extracted Text in Glide.
Click away to save your configuration.
A note on page selection
You may select a range of pages separated by a comma (for example, 2-3, 5, 7-9 will extract text from pages 2, 3, 5, 7, 9 ) or use * to select all pages. Google Cloud Vision limits the number of files at one time to five.
Extract text from image
The Extract text from image feature will extract text from an image. For example, you could extract the text from a picture of a receipt:
The action can be added to a Component that supports Actions (e.g., a Button component), added to an action sequence in the Action Editor, or added as a Computed Column.
If you have not yet set up a URL column to store the image file links, and a text column to store the extracted text, do that first in the Data Editor.
Select the component you'd like to add the action to. Remember, this must be a component that supports actions.
Search for the Extract text from image action or navigate to Integrations -> Google Cloud Vision -> Extract text from image.
Name the action.
Choose an icon.
Select the column with the File links.
Choose the language ID for the text. A full list of languages and IDs can be found here.
Choose the model, Document or Text.
Specify where to store the extracted Text in Glide.
Click away to save your configuration.
To learn more generally about Integrations in Glide, including how they affect your app’s usage, check out our Introduction to Integrations.