I recently went along to a University of Liverpool symposium showcasing case studies of generative AI in teaching, learning and assessment. The presenters and audience were predominantly from an HE background but the content, augmented by the suggestions from other educators in the chat provided some great transferable ideas.

The focus was on multimodal learning, which is academic speak for multimedia, apparently multimodal has more emphasis on the design and process (but when has a good educator ever really preferred style over substance!?) 

The big timesavers with the most impact were in using generative AI tools to speed up creating video scenarios and quiz sets. There was also some interesting work using ai avatars as personas for role-playing real-life situations. The team is currently working on a document to showcase all of the ideas presented which I will add to the toolkit as and when it becomes available.

Using genAI in video scenarios.

The example used was conflict resolution in working groups. The activity was to watch videos of each character in the group explaining the issues or difficulties they had with group work and then answer multiple choice questions based on the video.

First off ChatGPT was asked to create the persona for each student in the group;

“invent four persona descriptions, one for each member of a diverse group of engineering students tasked with working together on a joint assignment to design a new school building”

Next ask for a script

“Write a short script for Amina explaining that she feels the other students don't listen to or value her ideas”

Next ask for multiple choice questions based on the scenario in the script.

“Create 4 multiple choice questions about how Amina and the group handled the situation”

The next step is to turn the script into a wav file by using AI to produce a voice for each character. This was done using https://elevenlabs.io/ someone in the chat also suggested https://www.narakeet.com/

Next to make a video - Adobe express (free version) was the tool of choice for creating an avatar of each character and then adding the voiceover and animating the character. I’ve just had a quick look at Adobe express and it seems straight forward enough, Animaker is a bit easier to use and does the same job, someone in the chat suggested https://invideo.io

The video can then be uploaded to your Moodle or other LMS site. 

Then you can use h5p or genially to add the multiple choice questions.

If you work through this you can find the example https://liverpool.h5p.com/content/1292415699960858247

If you are usually the one filming and editing the video then this process could save you hours and is well worth investing a bit of time to learn how to do it.

More case studies and resources from the event are available via University of Liverpool Centre for Innovation in Education and their very useful Symposium Padlet