From a readable screen to a vertical Short
Inspect a narrated sequence of actual converter states, fitted into a vertical render with captions. The source is controlled demonstration material.
Inspect the result
- We captured three states of the real Hindi-to-Hinglish text converter, with original text and enlarged recording typography.
- Those states were assembled into a source video with synthetic English narration; this is not a continuous interaction recording.
- TheMomentra fitted the source into a vertical canvas and rendered the supplied narration captions. The complete source is retained, so small UI text still needs a phone-size check.
Download the exact output MP4 · Open the source video
Try this workflow in your own project
What this example does not prove
- Controlled UI states and synthetic narration, not a customer recording.
- This example uses full-frame fitting, with no presenter or automatic speaker detection.
- Fitting a wide screen can make text small. Enlarge source text, simplify the view or choose a closer source before exporting.
Script and caption timings
This is the script used to synthesize the source, with the supplied phrase timings. It was not recovered by an automatic speech recognizer.
| Source time | English |
|---|---|
| 0:00–0:04 | This is the real browser based Hindi to Hinglish text converter. |
| 0:04–0:10 | The original demonstration text keeps English terms such as API and YouTube. |
| 0:10–0:15 | Always review names and spellings before using the text in captions. |
Reproducible evidence
Rendered with TheMomentra on 27 October 2026 using eSpeak NG synthetic audio and the Noto Sans Devanagari font. No customer content or paid model API was used.
- Output
- 720 × 1280 · 15.30 seconds
- Output SHA-256
7bbef65e4eeaa343588b0f6b8f86bd8163dbe3fdadba06a1b821c968c1d62f95
Inspect the source script, render settings and artifact hashes · Read asset provenance