More on Google slides and Rachel Lucas' prompt for ChatGPT "Create a google slide size comic strip image to explain ser vs estar"

When Joe Dale publishes anything, I sit up and try it. So when he published a post about Rachel Lucas' YouTube channel , where she said she had produced three Google slide size images in three minutes, I had to try it myself.

Gemini

I was suspicious that I wouldn't have access to ChatGPT's image-generation features, so I decided to try the same prompts in Gemini. But what I got from Gemini was quite a different thing. It was a slide deck!

Gemini said, "Your slide deck on Ser vs. Estar is ready! I've designed it in a bright, vibrant comic book style to make learning these fundamental Spanish verbs as engaging as possible. Feel free to take a look and let me know if you'd like to make any edits." It offered the possibility to 'Export to Google Slides', which I took and once in Google Slides I couldn't resist the temptation to 'Convert to Video' To get a [British Voice] I added the instruction in square brackets at the start of every slide. I then downloaded two versions: one with music and one without. This is the version with music:

I had to remove the music as I got into trouble with YouTube for using  this music
"Uplifting and Inspirational" by

SharpFourStudio

And here is the version without music, which I prefer:

As I hadn't really wanted a slide deck, I decided to explain to Gemini what I wanted more precisely. I said, "Can you condense it into one slide image?" Gemini outlined what it would do, but I simply uploaded Rachel Lucas' slide and said, "Something more like this,"

With complete disregard for copyright it simply copied the example I had provided and 'produced' this:

I then asked Gemini "Can you do something similar to explain reflexive verbs?" and once Gemini had got the idea it produced this:

These two images were created very quickly, but with my third prompt, "Can you do something similar to explain noun and adjective agreement?" copying Rachel Lucas, the whole workflow ground to a halt. I had time to have a coffee, read the newspaper and it still hadn't finished. So, I asked, "Is there a problem?" "I'm still waiting for the last image"

Gemini replied, "I apologize for the delay. I am working on the image to explain noun and adjective agreement, but it is taking longer than expected. To ensure I deliver the best result, could you please provide some of the specific examples you'd like to see included? In the meantime, here is a general visual summary of the core concepts of noun and adjective agreement to help get you started." and provided this:

I complained, "There are some strange extra letters after some of the words like perroo, árboll, casaa, la mujerr felicees"

Of course Gemini apologised and eventually produced this:

There are still some problems with double letters in 'hombres felicees' and 'mujeres felicees'

So I gave up!

Back to ChatGPT

I wanted to see if by any chance I could generate the same sorts of comic strip images as Rachel Lucas did, so I prompted ChatGPT, "Create a google slide size comic strip image to explain ser vs estar" and got this:

Prompting, "Create a google slide size comic strip image to explain reflexive verbs" gave me this:

But the third attempted image creation gave me the dreaded, " Wait until 18.00 tomorrow or upgrade" message. I came back to ChatGPT the next day and prompted, "Create a google slide size comic strip image to explain noun and adjective agreement" by which time it had forgotten that the previous prompts had been about Spanish! I corrected the prompt to, "Create a google slide size comic strip image to explain noun and adjective agreement in Spanish" and finally got this:

There was one problem with the text and as I noticed that there was an 'Edit' option I used this and explained, "2 Gender - two collurds should be 2 Gender - two colours" and it worked. I got this:

It was a fairly long rabbit-hole to go down, but still probably worthwhile as I now know that Gemini and ChatGPT can both generate comic strip images, but they each need a bit of patience and guidance.

I discovered that ChatGPT allows you to edit images rather like in NotebookLM by explaining what you want to change.

I was more impressed by Gemini creating such a complicated Voiceover for my slides, when converted to Google Slides and then converted to Video. The British Voice was hilarious!