Vidrion Free

YouTube publishing pack

Your video already said it.
Vidrion writes the pack.

Drop in your video, its transcript, or a link — Vidrion builds titles, a structured description, validated chapters, hashtags, tags, thumbnail text and more, from the words actually spoken in your video. No sign-up. Your media file never leaves your device.

Drag & drop your video or audio file here

or click to choose a file — MP4, WebM, MOV, MP3, WAV, M4A…

Private by design — your file never leaves this device; transcription runs in your browser (one-time ~43 MB model download).

  1. 01Add video
  2. 02Transcript
  3. 03Pack
  4. 04Copy & publish

Your publishing pack

Cards follow the same order as YouTube Studio's upload form. Every card has its own copy button — take the whole pack, or only the pieces you need.

Start here — the core pack

Titles 5 angles

Add a transcript to generate this.

Description

Hook first — the opening lines are what viewers and search see.0 / 5,000

Chapters validated

Add a transcript to generate this.

Everything else in the pack

Hashtags

Add a transcript to generate this.

Only the first 3 appear above your video's title — they are marked Featured.

Tags minor role — YouTube says so itself

Comma-separated, ready to paste into Studio.0 / 500

Thumbnail text 5 options

Add a transcript to generate this.

Pinned comment

Community post

Shorts pack 3 moments from your transcript

Add a transcript to generate this.

Summary & search phrases

Add a transcript to generate this.

Readiness checklist

  • Title copied
  • Description copied
  • Chapters valid
  • Hashtags copied (3 featured)
  • Thumbnail text copied

Items tick off as you copy each part of the pack.

Or download one part:

Transcript the source everything above is built from

No transcript yet.

How it works

01

Add your video

Upload the video or audio file, paste a transcript, or start from a YouTube link. Three separate doors — Vidrion always tells you which one you came through and what it can honestly do with it.

02

Get the transcript

Uploads are transcribed on your own device by an on-device speech model — nothing is sent to a server. Pasted SRT/VTT keeps its timestamps; plain text is used as-is, and chapters are honestly skipped.

03

Build the pack

Titles, description, chapters, hashtags, tags and the rest are drafted from your transcript's own words and topics — chapters are checked against YouTube's stated rules (first at 00:00, at least 3, each at least 10 seconds) before you ever see them.

04

Copy & publish

Copy each piece into YouTube Studio, or export the whole pack as one file. Vidrion stops at ready-to-paste — publishing stays in your hands, in Studio, where it belongs.

What Vidrion will not do

  • It will not invent a description from a bare link — a link is metadata, not your video's words.
  • It will not promise views, ranking or “viral” anything. It prepares your pack properly; results are YouTube's and your audience's.
  • It will not upload, store or keep your file. Close the tab and the pack is gone — download it first.

Questions, answered plainly

Is my video uploaded to a server?

No. Transcription runs inside your browser, on your device. The one download involved is the speech model itself (about 43 MB, from Hugging Face's public library) the first time you transcribe — it is cached afterwards. Your video, audio and transcript are never sent anywhere by Vidrion.

Why does the link tab give me less than the upload tab?

Because a link can only show public details — the video's title, channel and thumbnail. Nobody can hear a video through its link, and tools that generate a “full analysis” from a link are making it up. Upload the file or paste the transcript and the pack is built from what your video actually says.

Which languages does transcription support?

English in this version — the on-device model is an English speech model. A pasted transcript can be in any language for most of the pack, but chapter and title drafting works best with clear, punctuated text.

How long a video can I use?

Your device does the work, so it depends on its memory. Files up to about two hours are accepted; long, large files are much happier on a desktop than a phone. If a file fails to decode, converting it to MP3/WAV audio and trying again usually fixes it.

Will YouTube accept these chapters?

Chapters are validated against YouTube's stated rules before you copy them: the first chapter starts at 00:00, there are at least three, and each lasts at least 10 seconds. If your transcript can't meet the rules, Vidrion says so in plain words instead of handing you chapters that would silently fail. Final acceptance is always YouTube's.

Do I need an account? Is anything stored?

No account, no sign-up, no server storage. Everything happens in this browser tab. Download the pack (or copy it) before closing the tab.