# Speaker-labelled captions

Edit / Edit effect / Captions. Checked 2026-11-13.

A muted viewer can tell who is talking.

- Use when: Dialogue, interviews, any two-voice format.
- Phrase: a small name tag above each caption line, one colour per speaker, the spoken word in a highlighted pill
- Page: https://directorchair.ai/effect/speaker-tag-captions
- JSON: https://directorchair.ai/api/effect/speaker-tag-captions.json

## Watch out

- In the measured two-voice reference 243 of the visual changes sit in the caption band, against 49 in the call tiles and 46 in the shared screen. The captions carry the rhythm; the picture changes about once every 7 seconds. (measured; Director Chair measurement of a published short, 23 Oct 2026)

## Prompts

### Any chat model

```
Give me an edit brief: where it goes on the timeline, the exact settings, and what to check before export, for a short video about [your product], for [your audience]. The technique is speaker-labelled captions: a muted viewer can tell who is talking. How it works: a small name tag above each caption line, one colour per speaker, the spoken word in a highlighted pill. It is right for: dialogue, interviews, any two-voice format. In the measured two-voice reference 243 of the visual changes sit in the caption band, against 49 in the call tiles and 46 in the shared screen. Be concrete: numbers, seconds and settings, no theory. Ask me one question first if something you need is missing.
```

## Goes well with

- Mock client call: https://directorchair.ai/preset/edit-mock-client-call
- Phrase captions on a chip: https://directorchair.ai/effect/caption-chips

## Source

Own words from Director Chair measurement of a published short, Oct 2026

Text CC BY 4.0, Director Chair (directorchair.ai). Reference clips and their card texts belong to their makers.