Behind the Craft by Peter Yang

Behind the Craft by Peter Yang

Share this post

Behind the Craft by Peter Yang
Behind the Craft by Peter Yang
My Favorite AI Tools to Create Images, Songs, Videos, and Avatars in 10 Minutes or Less
Copy link
Facebook
Email
Notes
More
AI Track

My Favorite AI Tools to Create Images, Songs, Videos, and Avatars in 10 Minutes or Less

Turn your ideas into creations in 10 minutes or less using simple text and images

Peter Yang's avatar
Peter Yang
Dec 03, 2023
∙ Paid
61

Share this post

Behind the Craft by Peter Yang
Behind the Craft by Peter Yang
My Favorite AI Tools to Create Images, Songs, Videos, and Avatars in 10 Minutes or Less
Copy link
Facebook
Email
Notes
More
6
6
Share

Dear subscribers,

Today, I want to share how you can turn your ideas into AI creations in just 10 min.

I think everyone should be tinkering with AI tools in their free time.

Doing so will help you understand what this new tech is capable of. It’s also a great activity to do with your family. So let’s cover how you can use AI to create:

  1. Text to image

  2. Text to song

  3. Text to video

  4. Image to video

  5. Text to avatar

All of these tools are free or have a free trial. I hope that you enjoy trying them out!


Text to image: ChatGPT and Clipdrop

ChatGPT

Let’s start with an old favorite — ChatGPT.

If you want a good laugh, here’s how you can get more from ChatGPT’s image feature:

  1. Type a prompt like “Give me an image of a product manager sweating.”

  2. Now repeatedly ask it for more (e.g., “More sweat!”)

The images that it generates will crack you up:

Image

Here are some other prompts that generate funny results:

  • “Create an image of a cute bunny.” → “Make it happier!”

  • “Create a spicy ramen.” → “Make it spicier!”

  • “Create a dad.” → “Make him more dad-like!”

Clipdrop

ChatGPT takes 30 seconds to generate images. Clipdrop generates images instantly as you type — perfect for kids with low attention spans!

Try it as a rainy day’s family activity.


Text to song

Riffusion

Riffusion’s text-to-song tool is a joy to play with:

  1. Use ChatGPT to generate lyrics (e.g., “Give me funny aggressive rap lyrics about how difficult being lactose intolerant is.”)

  2. Paste a verse into Riffusion and describe the sound that you want (e.g., a rap song with hard-hitting beats.”)

  3. It’ll generate a hilarious song you can share with others.

My lactose-intolerant rap song will top the charts soon

Text to video

Runway

Runway is the leading AI video tool but results are a bit hit or miss.

To get good output, try adding these keywords:

  1. Modifiers: masterpiece, classic, cinematic

  2. Camera: Full shot, close up, wide angle, slow pan, zoom

For example, here’s a video that I generated with the prompt:

A close-up shot of a cute corgi smiling in a cinematic, slow-zoom style

Pika Labs

Pika Labs is another popular AI video tool. Their latest model is behind a waitlist but you can use the current model for free by joining their Discord:

  1. Go into Pika’s Discord and pick a #generate channel

  2. Type “/create (your prompt)”

Here’s a video that I created using the same prompt as above.

If you’re struggling to get good results from text-to-video, try the image-to-video instructions below. I’ve noticed that it’s often more reliable.


Image to video

Runway

Runway has a motion brush tool that lets you animate a specific part of an image:

  1. Upload an image to Runway and add a text prompt

  2. Click on Motion Brush and paint the area that you want to animate

  3. Assign a motion direction and speed (horizontal, vertical, proximity)

The result is a really cool video like this:

Pika Labs

Pika Labs also supports image-to-video as follows:

  1. Go into Pika’s Discord and pick a #generate channel

  2. Type “/animate (your prompt)” and upload an image

Here’s Pika Lab’s video for the same image:


Text to avatar

Let’s get even more crazy!

HeyGen lets you make an avatar that looks and speaks just like you. To give it a try:

  1. Create a HeyGen account and tap “Free Instant Avatar”

  2. Record a 1-2 minute video of yourself talking about any topic

  3. Upload your video, give consent, and wait 5 min for your avatar to generate

  4. Type in a text script and then submit it to watch your avatar speak

Here’s an avatar of me talking about this newsletter with a British accent:

It doesn’t look quite right 😅 Nevertheless, in China, creators are already using HeyGen avatars to make money 24/7 from live shopping.


Here are all the tools that we covered above:

  1. Text to image: ChatGPT, Clipdrop

  2. Text to song: Riffusion

  3. Text to video: Runway, Pika Labs

  4. Image to video: Runway, Pika Labs

  5. Text to avatar: HeyGen

Now go out there and have fun creating with AI!


Enjoyed this post? Give it a ❤️ so I know to write similar posts in the future.


BONUS: A list of 25+ AI tools by category

If you want to explore further, here’s a list of 25+ AI tools organized into 7 categories:

  • AI video

  • AI image

  • AI voice cloning

  • AI translation

  • AI music and sound

  • AI avatar

  • AI virtual effects

  • AI video editing

The list below is available to paid subs only. Thanks for supporting my writing!

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2025 Peter Yang
Privacy ∙ Terms ∙ Collection notice
Start writingGet the app
Substack is the home for great culture

Share

Copy link
Facebook
Email
Notes
More