Saturday, July 05, 2025

Modifying the generated description to suit students' language levels

 One teacher, Anna Hung, on seeing my last poll, asked Perplexity to modify the generated description as follows: 

This is what I got using Perplexity AI, with the prompt: Describe the image above using simple grade 4 English for an esl class
This is a busy shopping street in Japan called Nakamise-dori. It has many small shops on both sides with red lanterns hanging. People are walking with umbrellas because it's raining. The shops sell souvenirs and food. The buildings have traditional Japanese roofs. Many visitors are shopping and looking at the stores.
Do you like visiting markets or shopping streets when you travel?

++++++++++++++++++++++

When I asked it to tell me in Grade 2 English:
This is a street with many shops. People are walking and holding umbrellas. It looks like it is raining.
What do you see in the shops? ☔
===========

When I asked it to use Grade 2 English to describe the buildings:

The buildings are long and low. They have red roofs and white walls. There are big signs and red lanterns on them.

Do you like the red roofs?

=========================================
Glad to see that we can have the generated description modified to suit the language levels of the students! I'm anticipating that there are different (creative) ways to capitalise on the 'Image > AI-generated description' process/product in the language classroom.

No comments:

Created a poem from a picture using Gemini: It did a super job

  昨天黃昏post了東涌日落我拍的照片,跟著Francis Tsang 提議我請AI 為之寫詩;這主意從未想過,於是立指打開Gemini, 上傳那照片,寫了一個簡單的prompt, 請它寫一首詩。 不用10秒,詩出現眼前,而且真的配合照片內容,連照片內細小的cable car ...