Derived formats with Gemini
Anand S
The natural capability of Generative AI is to generate stuff - and Gemini's particularly good with media.
For example, we can take any document, like this MasterCard report on The State of Open Finance 2026, and generate videos, podcasts, sketchnotes, songs, and more from it.
How?
I uploaded the PDF to NotebookLM and created a 20-minute podcast by clicking on Generate Audio Overview - Deep Dive - English - Default.
Audio: Listen to audio
It supports multiple languages, so I generated a Chinese and Filipino version as well.
Audio: Listen to audio
Audio: Listen to audio
Clicking on Generate Video Overview - Cinematic led to this video overview:
Video: Watch video
There are other formats in which we can generate videos. The Cinematic format is new, and the list is growing.
It's not just NotebookLM that you can use to generate new formats. Gemini itself supports a variety of formats.
For example, I used my Gemini Sketchnote prompt to create a visual summary of the report:
... and, using Lyria via the "Create Music" option to generate a narrative song with this prompt:
Create a narrative summarizing this article.
Narrate it rather than sing it.
Use a voice like Bobby McFerrin's, as if he were narrating rather than singing.
Keep the music minimal, focus on the voice.
Audio: Listen to audio
Next, I had Gemini create a slide deck by uploading the report and prompting:
Convert the attached report into a beautiful slide deck that conveys the most important actionable information for the audience.
STYLE:
Write it McKinsey style with action titles. Just reading the titles should give the audience the entire message of the deck.
Follow the pyramid principle. The contents of the slide should prove the title.
Make the slides content rich, i.e. clear and self-explanatory with enough detail to help the audience understand without a narrator.
Use iconography, typography, stock images, etc. as appropriate.
Write as a single page HTML application.
Then, a set of interactive explainers using this prompt:
Convert this report into 3 interactive explainers.
Pick the parts of the report that are best conveyed through interactive explanations. Identify the 3 most suitable ones.
Each explainer should, using animations, interactions, and simulations, explain a core point made in the report.
Render this as a single page HTML canvas.
Finally, a narrative data story using Claude -- which I could do with Gemini, too, but Claude is better at.
Where this is becomes practical is in:
- Proposals. No one pays attention to that company slide or RFP response. A 3-min video or 15-min podcast lets them absorb it during a walk.
- Reviews. Skip copy-pasting metrics into PowerPoint. Feed the raw data and ask for a McKinsey-style deck with action titles.
- Onboarding. Instead of a 100-page SOP or compliance manual, how about interactive explainers or a localized audio guide in Mandarin or Spanish?
- Manuals: How about a visual sketchnotes or step-by-step interactive flows from that documentation for call center agents?
- Case studies. Text-heavy fails. Maybe a 60-second narrative data story or sketchnote accompanied an upbeat narrative song?
- Reports. No one reads the 10-page competitor analysis. A 5-minute podcast or a single-page visual sketchnote helps the execs.
- Training. Create interactive simulations where people make actual decisions. Simsaram is my favorite example: family relationship training/simulation based on an iconic film.
- Emails. Why not use illustrations, sketches, flowcharts, etc. to liven up internal / external emails?
When generative AI makes generation easy, why not generate actually interesting stuff?