AI Transcription
Auto-transcribe a video into time-coded captions using the platform's configured provider (AssemblyAI, Deepgram, or Whisper). Runs in the background so long videos never time out — the editor shows live status.
AI Audio Dubbing
Generate a fully-dubbed video in any of 90+ languages with one click. ElevenLabs recreates each speaker's voice, tone, and pacing and keeps the background audio — download the rendered dub or save it back to Digital Assets.
Player-Synced Caption Editor
Edit every cue next to the playing video: fix the text, nudge in/out times, set a timecode from the playhead, add or delete cues, and watch the active caption highlight as it plays.
Configurable QA Standards
Define subtitling standards — characters per line, reading speed (CPS) target and max, min/max duration, gaps, and content rules — and the editor flags every violation live with error/warning badges. Set a workspace default and override it per project.
Per-Language Rules
Layer language-specific overrides on a standard — different reading speeds, quotation marks, decimal separators, measurement systems, and market guidance shown right in the editor for that language.
AI Quality Review
One click runs an AI pass that flags what rules can't catch — profanity, casual deity use, ALL-CAPS, literal or awkward phrasing, and off-tone lines — as advisory notes on each cue.
Any Video Source
Caption a video already in Digital Assets, upload a file straight into the app, or point at a video URL. The project links back to its source so footage and captions stay together.
Multi-Language Tracks
Keep one project per video with a separate caption track per language. Mark a default, and manage them all from one place.
Clean Exports
Export any track as WebVTT, SRT, or plain text on demand — cues are the source of truth, so exports always match your latest edits.
Spoken Audio
Turn a caption track into narration: pick a voice from the provider's catalog or describe the one you want, and the speaking rate is fitted to the caption windows so the words stay on the picture instead of finishing early. Mixed onto the video, with the original audio kept underneath if you want it.
Burn Captions In
Render the subtitles into the picture itself for the places that will not show a caption track — social, in-app players, a slide. Choose the size, position and style; runs on AWS Elemental MediaConvert.
Join Clips
Stitch several videos into one — an intro, a demo, an outro — trimming each clip, before you caption it. Doing it in this order matters: prepend a three-second intro to an already-captioned video and every cue in that track is three seconds early.
Save Anything Into Digital Assets
Every output — a caption file, a dub, a narration, a burned-in video — can be filed into the library: you pick the folder, make one if you need it, and name the file. Once it is there the rest of the platform can use it, including as the source of the next Video Hub project.
Embed-Ready Tracks
Each finished track has an unguessable public URL you can drop straight into a <track> tag on an external <video> embed.
Automate It
Every job announces itself, and both ends of it: project created, transcription completed, track imported, track exported, and the completion or failure of a dub, a narration, a burn-in or a join. Subscribe with a webhook or fire an orchestration workflow.
Drive It From an Agent
The whole sequence is available to the AI Assistant and over MCP — find a video in Digital Assets, caption it from a subtitle file already in the library, narrate it, and file the result back — with every step that writes something asking you first.