1. Word-by-word captions
Each word pops up alone in the middle of the screen at the moment it is spoken.
Show the prompt
Add word-by-word captions to input.mp4. words.json has the word timings. Show one word at a time in capitals, centered 70% of the way down, white bold text with a black outline, about 13% of the frame width tall, and only while that word is spoken. Keep the audio. Use Python with Pillow to draw the text and ffmpeg to put it on the video, and save it as output.mp4.
2. Highlighted word captions
Three words sit on screen at a time and the one being spoken turns yellow.
Show the prompt
Add captions to input.mp4 using the word timings in words.json. Show the words three at a time in capitals, centered 70% of the way down, white bold text with a black outline, and turn the word being spoken yellow (255, 212, 0). Each group stays up from its first word's start to its last word's end. Use Python with Pillow to draw it and ffmpeg to put it on the video (don't rely on ffmpeg's drawtext filter), keep the audio, and save it as output.mp4.
3. Punch-in zoom
The shot snaps 25% closer on the first word, holds for a second, then cuts back wide.
Show the prompt
In input.mp4, add a punch-in zoom at the start: zoom in to 1.25x over the first 0.15 seconds, hold it until 1 second, then cut back to the full frame. The zoom must stay centered: crop the same amount from all four sides, so the middle of the frame doesn't move. Keep the size, frame rate and audio the same. Use ffmpeg and save it as output.mp4.
4. Progress bar
A thin pink bar along the top edge fills from left to right as the video plays.
Show the prompt
Add a progress bar to input.mp4: a bar along the very top, 1.2% of the frame height, in (255, 59, 92) over a faint white track, filling from left to right so it is full on the last frame. Use Python with Pillow to draw it and ffmpeg to put it on the video (don't rely on ffmpeg's drawtext filter), keep the audio, and save it as output.mp4.
5. Lower third
A white name card with a red edge slides in from the left near the bottom and stays.
Show the prompt
Add a lower third to input.mp4: a white card with a red (230, 40, 60) strip on its left edge, showing "Jordan K." in black bold and "Verified customer" in grey under it, sliding in from off the left edge at 0.5 seconds over 0.4 seconds (fast then slowing), stopping 5% in from the left with its top 80% of the way down. Use Python with Pillow to draw it and ffmpeg to put it on the video (don't rely on ffmpeg's drawtext filter), keep the audio, and save it as output.mp4.
6. Freeze frame with text
The first frame holds still for 1.2 seconds under a headline before the clip starts moving.
Show the prompt
Make input.mp4 open on a freeze frame: hold the first frame for 1.2 seconds with "You need to see this" in black on a white rounded box near the top, then play the clip normally with the text gone. Delay the audio by 1.2 seconds so it stays in sync. Use Python with Pillow to draw the text box and ffmpeg for the rest (don't rely on ffmpeg's drawtext filter), and save it as output.mp4.
7. Picture in picture
A second clip plays in a small white-framed window in the top-right corner.
Show the prompt
Put inset.mp4 as a picture-in-picture over main.mp4: scale it to 32% of main's width, give it a 6-pixel white border, and place it in the top-right corner with a gap of 4% of main's width from the top and right edges. Keep main's audio. Use ffmpeg and save it as output.mp4.
8. Split screen
Two clips play at once, one filling the top half of the frame and one the bottom half.
Show the prompt
Stack top.mp4 above bottom.mp4 in one vertical video the same size as top.mp4. Each clip fills its half, cropped from the center. Use top.mp4's audio. Use ffmpeg and save it as output.mp4.
9. Text highlight marker
A yellow marker stroke swipes in behind the key words of the headline.
Show the prompt
Add a highlighted headline to input.mp4: "Clears dark spots" in black bold, centered 20% of the way down, and starting at 0.6 seconds a yellow marker rectangle swipes left to right behind "dark spots" over 0.4 seconds and stays. Use Python with Pillow to draw it and ffmpeg to put it on the video (don't rely on ffmpeg's drawtext filter), keep the audio, and save it as output.mp4.
10. Vertical blur background
A wide 16:9 clip sits across the middle of a vertical frame, with a blurred copy of itself filling the top and bottom.
Show the prompt
Turn the 16:9 clip input.mp4 into a 1080x1920 vertical video: the original sits in the middle at full width, and the space above and below is filled with a blurred, slightly darkened, zoomed-in copy of the same clip. Keep the audio. Use ffmpeg and save it as output.mp4.
