Using AI to Generate Graphics/Images

As I have posted earlier, this past summer at the American Classical League Summer Institute, I attended Stefanie Gigante's presentation "Artificial Intelligence (AI) for Latin Teachers." I have already addressed on this blog how she demonstrated using ChatGPT and Diffit to create texts but she also showed how Canva can be used to create AI-generated images and graphics.

I decided to try this out with a bellringer. Since it was the second week of school, I wanted to do some review of vocabulary words from last year with my Latin 2 students and to get them to write the words down with definitions, so in Magic Media, I typed in the following words in English as my parameters - I also set it to "graphics":

When I pushed "generate," after a few seconds, Magic Media gave me a number of possible graphics which included visual representations of the words. Not every graphic used every word, so I had it regenerate new graphics until I found one that I liked. Below is the image, along with how it was shown for that particular day's bellringer:

On the next day, I reused that graphic for a bellringer, but now the focus was on writing sentences in Latin about the picture. 

Observations

1. I love how I was able to use AI to create custom-made graphics. Instead of having to search for images/graphics which would fit what I want, I can have AI create them for me!
2. The more random the picture, the better I feel for an activity like this! 
3. Magic Media on Canva can create pictures and videos in addition to graphics.

What has been your experience in using AI to generate images/graphics?

Multiple Picture Listening Activity

This is a great post-reading, listening activity which I just recently learned from Andrew Olimpi, and it is very similar to Picture A/B Listening Matching but just with more pictures. This does require that you use pictures/screenshots so there is definitely some prep before.

Pre-Activity

1. Create 8-10 pictures/screenshots which illustrate sentences from a reading.
2. Place 3-4 of them on a Google Slide/Powerpoint.
3. Label each picture with a letter AND then include an additional letter choice as "None of the above"
4. Write sentences or use sentences from the reading which describe what are in the pictures AND write some sentences which do NOT describe the pictures. Six sentences is a good number.
5. Do the same for 2-3 more slides.

Activity

1. Project the first slide.
2. Read the first sentence and tell students that they are to write down which picture has been described. Let them know that "None of the above" is an option.
3. Continue reading sentences and have students write down their choices.
4. After the first slide is finished, review answers.
5. Continue the same with the second and third slide

Example from Latin 2 last week:

Sentences (with answers)

1. Gaius ad templum Iovis advenit ad implorandam victoriam in Circo Maximo. C
2. In templo Iovis erat magna statua dei. A
3. Noster parentes semper ad hoc templum advenerunt ad imprecāndam victorīam Romanam contra hostes. D
4. Cum Gaius vidisset statuam Iovis, exclamavit A
5. Cum noster parentes essent victorīōsi in pugnando bello contra hostes, advenerunt ad hoc templo ad honorandum te. B
6. In muris, Romani scribunt, “Scorpius est optimus!” et picturas Scorpi pingunt. D

Observations

1. This activity actually went very quickly, and I was able to do three slides of six sentences each (18 total) in about 10 minutes. 
2. I like the "None of the above" option, because students had to listen carefully and to look at the pictures closely to see if that was the answer.
3. Students need to be somewhat familiar with the pictures beforehand so that they know what the pictures are communicating, I had used these pictures the day before in a digital Picture/Sentence Matching assignment.
4. I used this as a practice for the next day's Picture A/Picture B Listening Assessment. (formative). Because students were very familiar with listening/picture matching aspect and had practiced the sentences earlier but in a different way, they did quite well.

Thanks, Andrew, for this activity!

Emoji Picture Story Retell

A few months ago, I wrote a post about using emojis as part of a Find the Sentence activity. Here is an extension activity which you can do with emojis and can be used following doing Find the Sentence. 

Today in my Latin 2 classes, I used emojis as a story picture retell. Yesterday, I had students do an emoji Find the Sentence with a Movie Talk reading which we had been doing for the past few days. Today, I gave students the story written in emojis (and some English words where there was no fitting emoji) and had them retell the story to each other in Latin in partners. The Movie Talk story itself was incredibly basic with lots of repetitions and limited vocabulary, so it seemed like a perfect story to "emoji-ize" and to experiment as a story picture retell.

Monsterbox

Observations

1. Because yesterday the class had done a Find the Sentence activity with these same emojis, students were already familiar with what Latin words these emojis represented, because meaning had already been established.
2. Because vocabulary was limited, it was a very easy story for students to retell relying only on the emojis. If the story had extensive vocabulary, I think that it would have been more difficult due to an overabundance of emojis.
3. Not every story lends itself to being "emoji-ized" due to not every vocabulary word having a matching emoji. As you can see in my story above, there are no emojis indicating size, so I had to write those words in English, as well as the verb want. 
4. As an extension, I had students then use the emoji story as a guide for a timed-write so that what they verbally expressed had a place to go.

Overall, using emojis in this way is a novel way to do a story picture retell, and it is definitely one that I will do again in the future. At the same time, however, it does have its drawbacks due to a limited emoji language.