Gemini getting better and better in generating images: Image Consistency, and Spelling

 I still remember not long ago, Gemini would sometimes display 2 problems when generating images: (1) image consistency of the main characters in a picture sequence, and (2) spelling. 

Today, those issues seldom occur again. Good job, Gemini.

Repicating someone's created comic strip and turning the dialogues into English: Good Results

Original Comic Strip by Ms Selena Tam: 

WHAT I DID: 

1. I went to Gemini > Image.

2. Paste in the above comic strip. 

3. Wrote the prompt: 

Replicate the uploaded comic strip, turning the Chinese dialogs in the speech bubbles into English

4. output below: 

================================

TAKEWAY: 

1. Gemini replicated the original images, without changing them in any way. 

2. The English translation from the Chinese is accurate. 

3. If teachers find a suitable comic strip in Chinese which they want to use in an English lesson, they can follow this example.  

 

Replicated a quest app on Food Pyramid and getting the share link using Netlify

 The entire original quest is developed by Miss Lily for a P4 five-lesson AI-enhanced unit on Healthy Eating. 

Her original quest app:

https://vast-canyon-ba7a.here.now/

===================================

I wanted to try creating something like her Lesson 1.

 So I took a screencap of Lesson 1, went to Gemini > Gemini Canvas, and asked it to create a similar webpage. 

It worked; but the resulting Share Link did not work. 

- I then saved the resulting webpage as an HTML file (which worked)

- I went to netlify, and dropped the HTML file into the upload box. 

-  It created a URL; I asked it to make it public, and got the below link: 

https://mellifluous-daifuku-594522.netlify.app/

===================

TAKEWAYS

1. By uploading the image of an interactive app to Gemini Canvas, it can replicate a similar one. 

2. Netlify can give you a public link to an html file.  

Gemini + Google Drive > Locating files under a topic and creating a quiz based on those files

 I then tried Ha Sir's second example in his explainer video (https://www.youtube.com/watch?v=qPAFfcOiJ14). In Gemini, I typed @, then chose Google Drive. I then asked Gemini: 'Please find my google slidesdecks which are related to language awareness.'

It took Gemini less than 10 seconds to locate, in my Google Drive, the names of the 6 Google slidedecks related to Language Awareness. I then continued to ask Gemini to create a 10-item MC quiz based on the 6 located slidedecks. Again, Gemini gave me the quiz, in interactive format (https://share.gemini.google/ZK1xd6n8KDDq)

Gemini + Google Services > First attempt to add events from Faculty Almanac to my Google Calendar

 Following Ha Sir's first example in his explainer video (https://www.youtube.com/watch?v=qPAFfcOiJ14), I uploaded the 26-27 faculty almanac (in Chinese) to Gemini, typed @, and chose 'Google Calendar'. I then asked Gemini: 'Add the 學位教師教育文憑課程 events in September from the attached faculty almanac to my Google Calendar.'

Ten seconds later, when I checked my Google Calendar, the 3 September events on the Faculty Almanac were there already. 

Gemini 可以聯繫其他的Google service (Keep; Gmail; Docs; Tasks; Google Drive), 那麽可以玩的事情便多了。

Saving a webpage as a desktop app: How

 STEPS

1. Go to the website that you want to install as a desktop app. 

2. Stay on that tab. Click on the three dots in the top right corner. 

3. In the dropdown menu, open 'Cast, save, and share'. 

4. On the next pop-up window, choose 'Install page as app.' 

5. Type out the name of the desktop use. Then click 'Yes'. 

6. The website is installed as a desktop app immediately. It can be (1) pinned to the task bar, and (b) listed in the installed app under 'Start'. 

Being surpassed by former students

 我們當教師的，最開心見到超越自己的是自己的舊學生。Kevin Wong 當年修讀我Master's Computer Assisted Language Learning(CALL) 一科，MA ELT畢業後遠赴美國深造，現時是加州Pepperdine University 的Associate Professor, 和Chair of the MA in TESOL Program. 他和Moorhouse去年出了運用AI 於英文教學一書，他現在CALL 的造詣肯定超越了我，但我只會覺得欣慰。

如果你也對運用AI 於英文教學有興趣，可以到下列Cambridge Univ Press網址免費下載全書的PDF版。(Click on the URL.)

Moorhouse, B. L. & Wong, K. M. (2025). Generative Artificial Intelligence and Language Teaching. Cambridge University Press. 

Created 2 interactive games using an uploaded infographic image using Padlet Arcade and Wordwall

 TOPIC: Multimodal Literacy and 21st century school education

APP: Gemini

STEPS: 

1. I asked Gemini to explain the significance of Multimodal Literacy for 21st century school education. 

2. Then I used the Image function, and asked Gemini to create an infographic based on its output. 

OUTPUT: 

=====================

CREATING INTERACTIVE GAME FROM THE ABOVE INFOGRAPHIC

I. Padlet Arcade

- After I uploaded the image, Padlet Arcade analysed it and then suggested a few game formats. I chose oe of them. 

https://arcade.padlet.com/game/yGjxwVOmr6?link_shared=1

II. WORDWALL

- I started by choosing a template (game format).

- I uploaded the infographic. 

- Wordwall created the game. 

https://wordwall.net/resource/116641088/multimodal-literacy-essential-21st-century-skills

Tried using Powerpoint Copilot to generate a presentation based on a Youtube video

今天嘗試在Powerpoint 內的Copilot,  給它一個youtube video link (Professor Li Wei 就Translanguaging 的mini lecture),請它以之生成一個presentation, 但Copilot 取得topic 後，自由發揮，以至生成的slidedeck 並不配合原來YouTube video的內容。

跟著我再試一次，這次向Copilot 聲明要跟足video 的內容，今次的結果便頗為滿意了。

當然更重要的是教學上如何使用基於YouTube video 生成的presentation? 暫時我想到的，就是遇到陌生或艱深的video 內容，學生純粹看video 未必跟得上，那麼便可以請學生先read through 生成的Powerpoint 一次，才觀看那video，這樣學習效果應該會更好。(當然也可請學生先閱讀YouTube video 的transcript, 或AI summary, 但Powerpoint slide 的視覺設計，效果應好過純文字。這樣先看AI 生成的Presentation, 才再觀看原來的video,這樣也算multimodal learning 吧。)

Converted a static worksheet to an interactive webpage using Gemini Canvas: Politeness Power

 STEPS:

1. I made a screencap of part of a worksheet (on politeness expressions) created by Gracelyn, and paste it into Gemini's prompt box in CANVAS. 

2. My prompt IN CANVAS: 

Convert the above task into an interactive HTML web application. There will be 2 subtasks: (1) Students will tick the sentence in each pair which they find more polite, and then (2) for the sentence which they consider more polite, students will click the word in the sentence which makes the sentence more polite.

3. Output 

https://share.gemini.google/CWupQdE1S1Rw

=========================================

TAKEAWAY: 

Gemini Canvas is doing a good job in converting a static worksheet into a web application. 

Created a three-tab interactive webpage on Simple Machines using Poe

 AI TOOL USED: 

- Poe's Canvas Creator

PROMPT

Create an app for upper primary students to learn about Simple Machines. Refer to the following picture for the format.

Learning Objectives:

• I can tell the names of the parts of a simple machine.
• I can label the picture of a simple machine.

There should be 3 tabs across the page below 'Learning objectives'.

• The first tab, titled 'Simple Machine', opens a picture with the parts of a simple machine labelled.
• The second tab, titled 'Quiz', opens up a few multiple choice questions about simple machines.
• The third tab, titled 'Simple Machine Labelling', opens up a diagram labelling task for students to label the parts of a simple machine.

OUTPUT:

https://poe.com/Simple-Machines

(Users need to sign in with their Poe account.)

Generating an infographic from a chat output: Direct in Gemini from an output vs Copy Output and generate in Image

 What I did: 

1. I asked Gemini chatbot: What is the relationship between Metacognition and Language Awareness? 

2.  I obtained the output from Gemini. 

3. While staying in the chatbot mode, I immediately asked Gemini to create an infographic based on the output. 

4. Below is what I obtained? 

I then tried another route. 

- I copied the earlier output from Gemini. 

- I went to Gemini Image, pastedd in the output, and asked for an infographic. 

- Below is the image output. 

==========================================

OBSERVATION

The first infographic looks better. It's easier to understand. The second one is a bit confusing, and too texty. 

TAKEAWAY

- Follow the first approach.