Start with your sound.
Load a recording, capture a performance, or record your voice. Choose the passage next, then add captions and shape the final output.
Clip the moment
Tune analysis · optional
Search transcript & create moments
Find a topic or name. Preview a passage, then add it to your moments to load and trim it.
Set a custom passage by time
People · name and match voices
Listen to a sample, name the person, and use “Same person as…” to link repeated voices. Rename once for all linked appearances. You can separate a mistaken match.
Paste a post (or any text). OutLoud narrates it in a real ElevenLabs voice and renders a feed-ready audiogram.
How it works: put your cursor where a new mood starts, then click a tag. It drops in right before the words it should color. No need to highlight anything (if you do highlight a sentence, the tag lands in front of it). The voice performs the tag; it's never read aloud and never shows on your captions.
[excited] We just hit fifty million streams! [whispers] I still can't believe it.
A tag carries until the mood shifts or the next tag. To settle back to a neutral read, start the next sentence untagged, or drop in [calm]. Tags need the Eleven v3 model; on other models OutLoud strips them so they're never spoken.
Voice expression· delivery & tags
Creative leans into emotion and audio tags · Robust stays the most consistent.
Design the moment
Make it yours. Set the format, shape the captions, and watch your changes live.
