If you have a paid subscription to Gemini (formerly Duet AI), you can generate images in Google Slides and Google Meet in a web browser. The feature is available to Google Workspace customers with the Gemini add-on (for $14 per user per month, paid annually for Gemini Business, or $30 per user per month, paid annually for Gemini Enterprise) and individual Gemini subscribers ($20 per month in personal accounts). Gemini in Google Slides or Meet offers an alternative to laboriously drawing custom images or selecting sterile stock photos; Instead, you can type text to describe the image you want.
As always, ensure that your use of AI-generated images complies with your organization's usage and attribution guidelines.
When using Gemini, Figure A shows how to create an image in Google Slides (i.e. Create Image with Gemini), and Figure B shows how to access the background image creation option in Google Meet (i.e. Generate a background). Activate the feature, enter text describing an image, optionally select a style from the drop-down menu, and then wait a few seconds for the system to generate images.
The drop-down style menu options differ between Google Slides and Google Meet. The Google Slides style drop-down menu defaults to No Style, but you can also select the options Photo, Background, Vector Art, Sketch, Watercolor, Cyberpunk, and I'm Feeling Lucky. While the I feel lucky option is a nod to one of Google's first search functions that automatically took you to a first result, in this case it allows the system to select a style. Similarly, the Google Meet background generator style defaults to No Style with Photography, Sci-Fi, Fantasy, 3D Animation, Illustration, and Film Noir options available.
How to review images generated by Gemini
You can review the generated images. When you select an image generated by clicking or tapping on it, the system adds it as a background in Google Meet or as an image in Google Slides.
If you are not satisfied with any of the generated images, select See more to try again. You can also edit the text message to describe the image you want differently. Google's Gemini support page suggests that you may get better results when your text describes the subject, setting, distance, materials, and background.
What types of images can Gemini create?
The variety of images that Gemini can create in Google Slides and Google Meet is enormous. To give you an idea of the variety and quality available, I generated five different types of images in different styles: an object, a scene, people, an idea and a sign. The images on the left below were the first four images the system generated, which I inserted into a Google Slide and then captured as a screenshot. The image on the right is a similar message used in Google Meet. Since the style options differ, the choice is indicated in each case below.
Generate an object
Without any style selected, the message “Laptop on a desk in an office” produced images that suggest a simple photograph of a common office scene on Google Slides and Google Meet (Figure C).
Generate a scene
A post of “Beautiful natural scene of a bird flying over the Rio Grande” resulted in an image on both Google Slides and Google Meet (Figure D) that represented a river with varying numbers of birds in flight. The watercolor style of Google Slides and the illustration style of Google Meet evoked the quality of works created by hand. Interestingly, the images generated in Slides included pictures within a picture: framed illustrations of the requested subject within the scene.
Generate an image of an abstract idea.
The message “Abstract illustration of a neural network” explored how the system could display a concept. The results were different, with Google Slides set in vector art style showing illustrations of neural networks, while Google Meet set in sci-fi style produced human faces tangled in network connections. (Figure E).
Generate an image with people
In my tests, the system sometimes refused to generate images with people. The message “Two people shaking hands, photorealistic” set to photo style in both Google Slides and Meet produced results (Figure F). These results are greatly improved over the starting hands produced in previous versions of Duet AI.
Generate a sign with text
Next, I tried to generate a request to generate a “Sign that says 'encourage experimentation'”, with the style option set to draw in Slides and fantasy in Meet, respectively (Figure G). While Duet AI had provided images in response to this message, Gemini's update results in a rejection. The system will not generate banners with text, so you will need to create them separately at this time.
Generate an image from literature.
When prompted for the wonderfully descriptive first paragraph of James Joyce's short story Two Gallants from his book “Dubliners,” Google Slides and Google Meet generated the following images (Figure H, left and right, respectively). Repeated trials often similarly produced only one or two images as a response, unlike almost all previous prompts that resulted in three or four sample images. The complexity of the text message likely affected the number of images the system could generate within a system-defined response time.
Mention me or send me a message at X (@awolber) to tell me how you use Gemini (formerly Duet AI) to generate images in Google Slides or backgrounds in Google Meet. What style cues and settings produce the images you prefer?
