Quickly creating a 'script + audio' set using Edcafe.ai: can be done very quickly.

 Just found a way to very quickly create a 'script + audio' set, within the same tool (Edcafe.ai). If you need to quickly generate a script and then its audio, without resorting to 2 different tools, then this workflow with Edcafe.ai is so far the quickest I have found. You can even assign the resulting listening clip/task immediately.

Here is my explainer page: https://workflowy.com/s/how-to-create-a-read/sAx91dz1HecsTJSD

Here is my created Audio asset: https://app.edcafe.ai/share/reading/tv-news-report-about-a-car-accident-6aa16fc6730fdc7f2e095b4c?cloneSig=5dfeb5ba326e84bae76a1a3559546381

There's one limitation though. Edcafe.ai can only create a monologue audio, not a dialogue. (The free plan is quite generous.)

Insert AI voice into a poster in Canva: steps

 1. Finish editing the image page. 

2. Click on AI voice on the left-hand pane. 

3. Choose a voice (eg, AmE vs BrE; male vs female) to preview. 

4. Then click 'Generate Voice' at the bottom. 

5. Then Edit as video. Adjust the length of the voice track. 

6. Download the video as MP4. 

7. The above can be repeated page after page. 

=========

TAKEAWAY

Can add voiceover to images in Canva. 

Tried creating a podcast overview using NotebookLM, result not useful

 This was my second attempt to use NotebookLM to generate a podcast from an uploaded document. I was thinking  of using the output for Flipped Learning  for ENLT6704, S4

I uploaded a Word document, which explained the relationship between metacognition and metalinguistic awareness (AI-generated). 

The first rendition was 18 minutes in length. 

I asked it to try again and limit the podcast to about 4 minutes. The resulting episode was 04:10 in length. 

THE BIGGEST PROBLEM: 

1. The podcast is a conversation between a male and a female. 

2. The biggest problem: It is like a chit-chat between 2 people, who chatted around the topic. They did not specifically mention the main concepts involved.

3. So, it serves the purpose of leisure listening, but can't be used for flipped learning. 

4. The result is very much unlike the video podcast, which is much more straightforward in presenting the main ideas. 

Notebook LM's fantastic Audio Overview function: You can join a generated podcast discussion

 NotebookLM 的generate audio podcast功能也很好玩，可以base on你upload 的material 生成podcast, 可以是單人的presentation, 但我更喜歡生成二人的討論，因為更lively和 interactive。

我聲明聽眾是 secondary-school ESL teachers in Hong Kong, 它在討論中也「整色整水」提及香港的英語教師。

它第一次生成的對話是用American English, 我請它用British English 再generate一次，它也照做。最精彩的，是它完成生成一個interactive discussion的podcast後， 聆聽時你可以join 這個discussion ，就是隨時加入，問問題和發表意見，而他們也似模似樣的回應！

Created a podcast episode about the latest ideas on using AI in ELT using Copilot Chat

 TOOL: https://copilot.microsoft.com/chats/

MY PROMPT: 

Create a podcast about the latest ideas on use AI in ELT in a casual tone

OUTPUT:

https://copilot.microsoft.com/shares/podcasts/H4Ufg2qhcuE39R7NMHNoo

====

OBSERVATIONS

1. In this attempt, I used a very brief topic as the prompt. Future attempts might use something more specific for the prompt. 

2. The generated episode is in the form of a dialogue between a man and a woman. Not sure if I can generate a monologue. 

3. For future attempts, could try customising the language level. 

4. Output audios can be shared via a link, but cannot be downloaded. 

Used Murf.ai to create a dialogue for presenting adverbs of frequency

 Earlier on, I used Quillbot AI Dialogue Generator to create a dialogue that contains 4 adverbs of frequency (Open English I, Unit 1). The dialogue is for presenting/revisiting the adverbs of frequency. 

Then I went to Murf.ai to create an audio dialogue from the dialogue text generated with Quillbot. 

Below is the Output: 

https://murf.ai/share/mfrzo8la

Murf.ai is good for creating DIALOGUES. IT also has a few parameters for users to customise their intended artifact (eg, speed; tone)

However, the Free Plan only allows for 10 minutes of AI generation. 

Also, the free plan does not allow downloading the audio dialogue. 

Tried using Google Gemini to create an interactive Listening game

 For this attempt, I used the same story text and prompt for creating the 10-sentence listen-and-match game. When in Google Gemini, among the 4 options inside  the prompt box, I chose 'canvas'. 

OUTPUT: 

https://g.co/gemini/share/a15aae917247

HOW IT COMPARES WITH POE AND CANVA AI

1. It works. 

2. The resulting task can be shared via the share link. 

3. There is no 'publish' vs 'preview' distinction. 

4. The resulting task works in Opera immediately. But in Chrome, users need to log in to their Google account first. 

5. The audio for each sentence takes a bit of time to follow. In this regard, it is not as good as POE and Canva AI. 

Created with Canva AI: A listen-and-match game for KS1 - Colin's story

 MY PROMPT: 

Please create an interactive Listen-and-Match game for lower primary ESL students to practise Listening. On the left of the page are 10 play buttons, numbered from 1 to ten. On the right of the page are the 10 sentences in random order. Students will click on the first button, listen to the sentence, which is said in British English accent, and match it to the correct sentence on the right. Then they click on the second button, listen to the sentence, and match it to the right sentence on the right. And so on until they finish trying the 10 sentences. Then show students the number of correct answers they have got. 

Below are the 10 sentences to use in creating this game: 

One morning, Colin, a six-year-old boy, went to the park.

He wanted to play on the swing.

But the swing went too high, and his hat flew off!

Colin ran after the hat, but it landed on a dog.

The dog barked and ran around with the hat.

Colin laughed and chased the dog.

Then he tripped and fell on the soft grass.

When he got up, he saw a butterfly on his nose!

The dog finally dropped the hat.

Colin went home smiling, with his hat and a funny story.

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

OUTPUT: 

https://paulsze.net/listening-game-colin-story

Created with Canva AI: A Listening Game for Phonics/Phonetics practice - Vowels

MY PROMPT: 

Create a Listening game. This game consists of 10 questions that correspond to the following 10 pairs of words. When each question is shown, the two words in each pair will be shown to students. One of the two words will be pronounced two times. Students listen to the pronounced word, and click the one which is being pronounced. Then the next question will show up, and so so. 

The 10 pairs of words to be used: 

sit, seat

cod, cord

bet, bat

pull, pool

men, man

cut, cart

live, leave

pin, pan

bull, ball

heart, hot

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

OUTPUT: 

https://paulsze.net/listen-and-choose-vowels