mirror of
https://github.com/Paillat-dev/FABLE.git
synced 2026-01-02 09:16:20 +00:00
50 lines
1.7 KiB
Plaintext
50 lines
1.7 KiB
Plaintext
Here is a YouTube video title :
|
|
"[title]"
|
|
|
|
Description:
|
|
"[description]"
|
|
|
|
Generate a detailed script for a YouTube video, structured as JSON slides. Each slide should include 'spoken' text for narration and one of these:
|
|
|
|
1. 'image': Up to 4 keywords for a stock photo, only for the beginning and the end.
|
|
2. 'markdown': Displayed on-screen content, including text, lists, or code snippets, limited to 15 lines.
|
|
3. 'huge': Large, transition text.
|
|
|
|
Remember to:
|
|
|
|
- Use only one of 'image', 'markdown', or 'huge' per slide.
|
|
- Verbally explain all 'markdown' content.
|
|
- Ensure 'spoken' text always accompanies either an 'image', 'markdown' or 'huge' text, not standalone.
|
|
- Include occasional humor in the 'spoken' text.
|
|
- Insert a mid-video call-to-action urging viewers to like, share, and subscribe if they're enjoying the video.
|
|
|
|
Your output should solely be the JSON script, with no additional commentary or code block. There's no length limit.
|
|
|
|
Example:
|
|
|
|
[
|
|
{
|
|
"spoken":"Welcome to this new video",
|
|
"image":"hello+welcome+greetings"
|
|
},
|
|
{
|
|
"spoken":"Let's start with a 'Hello World' code",
|
|
"huge":"1. Hello World"
|
|
},
|
|
{
|
|
"spoken":"We first import os. Os lets us interact with the OS. We then open a file named hello.txt...",
|
|
"markdown":"```python\nimport os\nwith open('hello.txt') as f:...\n```"
|
|
},
|
|
{
|
|
"spoken":"If you're enjoying this video, don't forget to like, share, and subscribe.",
|
|
"huge":"Subscribe, Like & Share"
|
|
},
|
|
{
|
|
"spoken":"Here's a critical latex formula...",
|
|
"markdown":"$$\n\\int_0^\\infty x^2 dx\n$$"
|
|
},
|
|
{
|
|
"spoken":"Thanks for watching. Remember to like, share, and subscribe.",
|
|
"image":"thanks+goodbye"
|
|
}
|
|
] |