Prompts

Podcast clip caption prompts

A podcast caption prompt sets the caption style, tells Vyra how to handle two speakers, and names the one line that should get a bigger treatment.

By Sulan Zhang, co-founder of Vyra · Updated 2026-09-24

Set the style, say how to handle two speakers, and name the one line that gets bigger. Podcast clips get watched on mute more than anything else, so the captions are the clip.

Minimal:

Add captions to this clip. Phrases, white, bold, lower third.

Standard:

Caption the whole 45-second clip in 3 to 5 word phrases, bold white with a thin dark outline, centered in the lower third. When the speaker changes, start a new caption on the first word of the new speaker so lines never mix two people. Keep captions clear of both faces.

Full:

Caption this clip word by word, TikTok style, bold sans-serif, white, sitting just above the bottom safe zone. Never let a caption group span two speakers. Break at every speaker change. When she says "nobody is going to read your newsletter", make those words larger, one word per beat, and drop the ordinary captions for that sentence, then resume. Add a small name label in the bottom left corner for the first 3 seconds each person is on screen, "Sulan" and "Caleb". No emojis, no pop animation.

What each part does

"Break at every speaker change" is the rule that keeps a conversation readable. Without it, captions group by timing and mix two voices in one line. The one big line is where the clip earns its stop. Name labels answer "who is this" for someone who found the clip cold.

Fixes

  • Two speakers in one caption. "Split every caption at the speaker change, even mid-sentence."
  • Captions over a face. "Move to the lower third and shrink 20%."
  • Wrong word. "At 0:18 it should say newsletter."
  • Emphasis not landing. "Make those words 2x and yellow, hold until she finishes the sentence."

FAQ

Word-by-word or phrases? Phrases for calm conversation. Word-by-word for a punchy claim. If unsure, phrases.

Should I show names? Yes, briefly, the first time each person is on screen.

Related

  • /prompts/podcast-clip-selection
  • /prompts/captions
  • /how-to/choose-captions
  • /formats/podcast-clip

Rules that apply to every prompt

  1. One target per prompt. Length, platform, purpose.
  2. Quote the exact line when you mean a specific moment.
  3. Give numbers, not adjectives. "Pauses over half a second," not "snappy."
  4. Say what not to add. Zooms, transitions, music, effects.
  5. One fix per follow-up message. Ask for a list of changes with timestamps if you want to check.

Try it on your own footage

Upload your clips, say what you want, get the edit back. Free to start.

Start editing with Vyra