The Operations Layer for Accessible Media
Closed Caption Unity is the missing operations layer for captioning and accessible media. Manage assets, run production, edit in Closed Caption Creator, publish house formats, and deliver to the platforms your clients already use — on your infrastructure, with your keys.
Production workspace
Everything your coordinators and operators need to see what is due, who is on it, and what ships next.
Dashboard
A live snapshot of open work, due orders, assigned tasks, recent assets, and active projects. Operators can rearrange widgets; org admins also see credits, storage, and team size.
Assets
The center of Unity. Each asset holds media, metadata, linked projects, jobs, and delivery history. Assets work even if you never create a project.
Projects
One asset, many jobs — for example English captions and French subtitles on the same source. Duplicate projects to reuse task structure and QC settings. Kanban view for task status.
Orders
Optional business layer: batch client work, track lifecycle (Open → In Progress → Review → Ready for Delivery → Closed), and deliver the whole order at once.
Customers
Client records (name, logo, contacts) so orders can be filtered and reported by account. Not an invoicing system.
Personal task list
A “what do I do next?” queue for captioners, translators, and QC. Assigned users can work a task even if they cannot see the parent project.
In-app notifications
Assignment and workflow alerts in the toolbar, so operators do not live in email.
Caption & accessibility editing
A professional editor built into the workflow — powered by Closed Caption Creator.
Embedded Closed Caption Creator
Full professional editor inside Unity: timing, positioning, speakers, QC, search/replace, and more. Keyboard shortcuts and editor preferences follow the user.
Auto-loaded resources
Proxy, waveform, first subtitle file, and transcript load when a task opens. Load Resource can add shot-change markers and other asset files mid-session.
Transcript sync
Check Sync / Resync align caption events to a transcript already on the asset.
Publish without leaving the editor
Choose an export profile; Unity converts and attaches the file to the asset. Publishing does not close the task, so editors can export as they go.
Caption QC in real time
Max CPS/WPM, min/max duration, max lines, max characters per line, overlap protection, and frame gaps — enforced while editing.
Audio description export
Kokoro TTS stitches cues into FLAC, MP3, or WAV. Voice-over or ducked mixdown, with optional EBU R 128 / ATSC A/85 / LUFS loudness.
Proxy review player
Browser playback with caption tracks so producers can review without downloading masters. Optional visual watermark (email, IP, date/time, custom text) on proxy players.
Ingest, processing & AI
Media arrives ready to work — proxies, metadata, and a first-pass transcript, with humans in the loop.
Direct-to-storage upload
Files go straight to your object storage (S3 multipart or GCS resumable). Unity does not proxy the bytes through the app.
Remote ingest
Pull from a public HTTPS URL or Google Drive.
Typed media files
Original, proxy, audio, thumbnail, transcript, subtitle, shot changes, waveform, attachments, and audio description — each can live in its own storage location.
Proxy & review media
Automatic 720p proxy, thumbnail, spritesheet, and waveform so editors are ready immediately.
Shot-change detection
Markers for the editor via FFmpeg or Google Video Intelligence.
Technical metadata
Duration, frame rate, drop-frame, resolution, and codec extracted on ingest.
Automatic transcription
Speechmatics (language auto-detect; SaaS default) or Deepgram. Timed transcript lands on the asset for editing and sync.
Automatic translation
Lara AI translates published subtitles into target languages and attaches the new files — with a human review step in the workflow.
Bring your own keys
Connect your own transcription, translation, transcode, TTS, storage, and platform credentials.
Formats & delivery
House formats in, finished files out — powered by Closed Caption Converter.
45+ caption formats
Closed Caption Converter handles SCC, SRT, WebVTT, TTML/DFXP/IMSC, MCC, EBU-STL, Cavena 890, iTunes Timed Text, and many more.
Export profiles (house formats)
Admins decide which formats editors may publish, plus conversion filters (reading speed, event gap, auto QC, 608/708, encoding).
Delivery profiles
Templates for “what files, which account, which platform.” Deliver from a project, an order, or the API.
Streaming platforms
YouTube, Vimeo, Mux, Brightcove, Bunny Stream, Cloudflare Stream — video plus captions (and extra audio / AD where the platform supports it).
Review & MAM
Frame.io (per-asset folder), Iconik, Telestream Vantage Cloud, Google Drive.
Object storage delivery
Copy required files to Amazon S3, Cloudflare R2, or Google Cloud Storage with prefix/filename control.
Teams, security & access
Give every captioner, reviewer, and freelancer exactly the access they need — and nothing more.
Organizations
Isolated tenants. Users belong to one organization.
Roles & workspace rights
Super admin, admin, user, and guest, plus per-workspace rights (assets, orders, projects, tasks, customers, upload, download, editor, and so on).
Teams & ACLs
Folder, profile, metadata, and order permissions by user, team, or everyone. Permissions are additive — the more-permissive grant wins.
Skill-based auto-assign
Attributes (for example “Spanish Translation”) on users and tasks automatically route work.
Guest task shares
Password-optional public links, optional expiry, email to Unity users or external addresses. Freelancers can complete a task without full product access.
Two-factor authentication
Authenticator apps (TOTP) and backup codes. Organizations can require 2FA for password users.
SAML 2.0 SSO
Enterprise / on-prem: Okta, Entra ID, or any SAML 2.0 IdP, with just-in-time provisioning.
Google sign-in
Available on the SaaS product.
Password policy
Per-organization length and complexity rules.
Encrypted credentials
Third-party API keys stored encrypted at rest.
Platform, automation & deployment
API-first and built to fit your infrastructure — in our cloud or yours.
Visual workflows
State machines with notify, create job, update metadata, and update delivery status. Ingest and delivery profiles attach to these machines. (On SaaS, platform workflows are seeded; on-prem admins can build their own.)
Custom actions
Right-click actions on assets and folders: open a URL, POST to an endpoint, or start a workflow (for example Automatic Translation).
REST API
The Workspace UI is a client of the same public API. Session or API key. Headless Unity is a supported model.
Webhooks
Subscribe to asset, project, order, task, ingest, folder, workflow, and delivery events. At-least-once delivery with idempotent event IDs.
MCP server
Connect Cursor or Claude to your Unity instance to search records, update tasks/orders, and trigger non-delivery workflows. Media stays on your instance.
SaaS or on-premise
Cloud-hosted with public signup and Stripe plans, or single-tenant / on-prem with your storage, SMTP, editor URL, and SAML.
Bring your own storage
AWS S3, Cloudflare R2, Google Cloud Storage, or local/NAS for on-prem.
Built on Three Principles
Human-in-the-loop AI
AI handles transcription and translation drafts. People QC, edit, and approve before anything ships — quality stays your call.
Modular by design
Orders → projects → tasks is there when you need it. Skip orders, or run Unity headless via the API — use only the pieces that fit.
Data sovereignty
Cloud or on-premise, your storage and your keys. Your media never has to live in a vendor silo you cannot leave.
Connected to Your Stack
Live integrations for delivery, storage, transcription, translation, and editing.
Mux
Brightcove
AWS S3
Google Cloud Storage
Speechmatics
DeepgramLaraSee It With Your Own Media
Start a 14-day trial with 2,000 credits, ingest a file, and deliver your first captions today.