Import
Choose a video and prepare its audio for caption timing.
Import a video, turn speech into timed captions, correct any word, shape the style and export a finished result — without turning a phone into a crowded desktop editor.
Datagram One is built around the shortest useful path, with manual control always available when the automatic result needs correction.
Choose a video and prepare its audio for caption timing.
Convert spoken audio into editable caption segments with timing.
Apply typography, colors, backgrounds, motion and placement.
Render the captioned video or export subtitle files for another workflow.
Automatic transcription is only the starting point. Every caption remains editable before export.
Generate caption text from the spoken audio in the selected video.
Adjust segment timing and fine-tune words when the automatic timing needs help.
Tap any line, fix a word, split a caption or merge segments without rebuilding the project.
Right-to-left layout and mixed-script handling are first-class design requirements.
Move captions vertically or freely while keeping social-video safe areas visible.
Control how many words appear at once instead of forcing every video into one caption pattern.
No locked template look. Presets are a starting point; typography and motion stay editable.
The preview stays central. Editing controls appear only when needed, with a clean timeline, large touch targets and a keyboard-aware correction flow.
The splash direction uses the ONE wordmark with a compact voice-wave motion, then lands directly into the workspace.
Planned output includes rendered captioned video plus subtitle files such as SRT and VTT. Social aspect ratios are part of the editor from the beginning.
Speech recognition can make mistakes when audio is noisy, speakers overlap or pronunciation is unusual. Datagram One will not present transcription as infallible; the editor is designed to make corrections quick before export.