For IT and administrators

The security review, on one page.

A desktop app that needs no account, no server and no analytics. Its data is one folder. It downloads only its speech model, the file it reads photos of a page with (if someone uses that), and its own updates. Everything it touches is listed below.

Security review, short version

Here is the short version of the security review. No account is needed, and there is no server to stand up.

Recordings, scripts and projects live in one folder on the device. Captions are made there too. The one download is a speech model, fetched once, and you can put it in place yourself. The rest is on this page.

170 wpm
The app's prompter, in small. There the highlight follows your voice; here it keeps a steady pace.
Accounts
None needed
Servers to run
None
Reports it sends
A crash report, if the user says yes
Where data lives
One folder per user
Downloads after install
The speech model, a photo-reading file if used, and updates
Programs it starts
Its own ffmpeg

Network

What leaves the device.

What the app doesWhere it happensNetwork
Records camera, microphone and screenOn the device, written to disk as it goesNone
Edits, renders and exports videoOn the device, with the ffmpeg it ships withNone
Makes captions and transcriptsOn the device, with Whisper (whisper.cpp)One download of the speech model, the first time
Reads a photo of a page into a scriptOn the device, with TesseractOne download of the language's file (3 MB for English) from app.lecternvideo.com, the first time it's used
Follows a voice, takes dictationThe platform's recognizer, or Whisper on the deviceIt depends on the platform: see speech engines
Goes liveOnly if someone chooses Go live on…: the bundled ffmpeg streams from the deviceRTMPS or RTMP, straight to the service chosen (YouTube, Facebook, Vimeo, BoxCast or another server), or, if someone chooses Through Lectern, to Lectern's relay, which sends it on. Stream keys are kept in the system's keychain
Posts to a platformSaves an MP4 and opens the platform's upload page in the browserNone from the app itself
Signs in to a Lectern accountOnly if someone chooses to (for Pro, sync, share links, or a seat in your organization's plan). Nothing needs oneHTTPS to Lectern's servers, and only after someone has signed in on that device
Syncs the libraryOnly when someone is signed in and has storage of their own: new recordings, scripts and projects, unless Sync new files is off, and what they choose to syncHTTPS to Lectern's storage, kept private to the account. Signed out, nothing syncs
Shares a video by linkOnly when someone signed in chooses Share link in PublishThe exported video, uploaded over HTTPS to Lectern's storage
Shares a script for commentsOnly when someone signed in chooses Share for commentsA copy of the script's text, over HTTPS to Lectern's cloud; reviewers' comments come back the same way
Uses another device (remote, prompter, camera)Only when someone signed in starts a session, and others join it by its linkSmall control messages through Lectern's cloud over WebSocket (WSS). A joined camera's picture goes device to device over WebRTC, found through Cloudflare's STUN server, or through Cloudflare's TURN relay where networks block a direct connection
Reports usageIt doesn'tNone
Reports a crashOnly if the user says yes when it asks, after a problem, or picked Send a report without asking in Settings. Don't keep or send reports turns it off. The report is shown before it's sentHTTPS to Lectern's cloud (app.lecternvideo.com/v1/crash-reports), which passes it on to Sentry without the user's network address. It holds what went wrong in Lectern's code, the app's version and the system: no files, paths, scripts or account
Updates itselfThe desktop app checks for a new version a few seconds after it opens and every few hours while it's open, downloads it in the background, and installs it when the user restarts the app. The phone apps are updated by their storesHTTPS to Lectern's cloud (app.lecternvideo.com/updates/), then the update itself, which is signed. The check sends nothing about the user

The one download

The first time Whisper is used (for captions or text from audio, and for voice follow and dictation on Linux and Windows, or wherever Whisper is chosen in Settings), the app fetches the Whisper base model: one file, ggml-base.bin, 148 MB, over HTTPS from app.lecternvideo.com (our own copy of the file on Hugging Face). If that can't be reached, the app tries huggingface.co, which hands the download on to its file host. Either way it's checked against a SHA-256 pinned in the app before it's used.

Or put it there yourself

On a network that blocks the download, place the file at speech-models/ggml-base.bin in the app's data folder. The app finds it and doesn't ask for it again. Until it's there, captions and Whisper's voice follow say they need it; everything else works.

Speech

Speech engines, platform by platform.

Voice follow and dictation listen to the microphone. Which engine does the listening, and whether it needs a connection, depends on the platform.

PlatformEngineWorks offline
macOS, iOSApple SpeechYes, when the language's on-device model is installed
AndroidThe system's SpeechRecognizerYes on Android 12 and later, with an on-device recognizer
Windows (default)Whisper, on the deviceYes, once the model is in place
Windows (chosen in Settings)Windows dictationNo. It needs Windows' online speech recognition turned on, and Microsoft's service has had outages
macOS (chosen in Settings)Whisper, on the deviceYes, once the model is in place
LinuxWhisper, on the deviceYes, once the model is in place
BrowserThe browser's Web Speech APIIt depends on the browser. Chrome sends audio to Google

Captions and transcripts of a finished recording always use Whisper on the device, on every desktop platform. Scrolling at a set speed uses no speech engine at all.

Data

One folder of data, one of cache.

PlatformDataCache
macOS~/Library/Application Support/com.lecternvideo.lectern/~/Library/Caches/com.lecternvideo.lectern/
Windows%APPDATA%\com.lecternvideo.lectern\%LOCALAPPDATA%\com.lecternvideo.lectern\
Linux~/.local/share/com.lecternvideo.lectern/~/.cache/com.lecternvideo.lectern/

In the data folder

  • library/ Everything a user makes: Recordings/, Scripts/, Projects/, Exports/ and Trash/ (Recordings/ is what the app shows as Media). Scripts are Markdown text, projects are JSON, media is ordinary video, audio and image files, and a recording's transcript and captions sit beside it. Once a user syncs, a hidden .sync.json records what's synced.
  • speech-models/ The Whisper model, once it's been downloaded or put in place.

In the cache folder

  • previews/ Thumbnails and waveforms for the editor's timeline. Safe to delete: they're made again when needed.
  • transcribe/ Audio decoded from a recording while it's transcribed.

Back up or redirect library/ and you have a user's work. The trash empties itself after 7, 30 or 90 days, as the user sets.

Permissions

What it asks the system for.

It asks forToScope
Camera and microphoneShow the preview and recordThe devices the user picks
Speech recognitionFollow the reader's voicemacOS and iOS ask; elsewhere it comes with the microphone
Screen recordingRecord a display, a window or a regionThe system's own picker chooses what's shared, each time
FilesKeep the library and the timeline's cacheThose two folders, and files the user picks in the system's open or save dialogs
Other programsConvert and render videoOnly the ffmpeg inside the app
Links and foldersOpen a web page, or show an export in its folderThe system's default browser and file manager
A local port (Linux only)Play the library's media in the web view127.0.0.1, a random port and token each run, library files only

Components

What's inside it.

  • Tauri 2, and the system's web view. WebView2 on Windows, WKWebView on macOS, WebKitGTK on Linux. No browser is bundled, so it's patched with the system.
  • ffmpeg, an LGPL build. No x264 and no other GPL parts. H.264 comes from the system's encoders (VideoToolbox, Media Foundation) or OpenH264.
  • whisper.cpp, with OpenAI's Whisper base model, which knows many languages.

Where it runs

  • macOS Apple silicon. Screen recording needs macOS 15 or later.
  • Windows With WebView2, which Windows 11 includes.
  • Linux With WebKitGTK 4.1 and the GStreamer plugins for camera and playback.
  • iOS, Android and the browser The phone apps do all but record the screen and post to sites; they aren't out yet. Planned The browser has fewer features.

Deployment

Rolling it out. Not all built yet.

You'll wantWhere it stands
Signed and notarized macOS buildsBeing set up Planned
Signed Windows installersNot yet: installers are unsigned Planned
Automatic updatesBuilt in: the desktop app checks for and installs signed updates
Silent install and MDM notesTo be written Planned
An accessibility conformance report (VPAT)To be written Planned

Questions

Asked before it's approved.

An administrator asks

Does it collect student data?

Not unless someone signs in. The app needs no account, and nothing a user records, writes or answers leaves the device unless they sync it, share it or go live. An account (for Pro, sync, share links, or a seat in your plan) holds a name, an email address, a password (stored hashed), the devices signed in with their network address, and what was bought; your plan's admins see members' names, email addresses and the storage they use. Anyone can download or delete their account from the app. Sync uploads a user's library to their own storage, where only they can open it. A share link uploads what it shares. Your plan's admins can turn off sync, share links and streaming through Lectern for everyone in it, so nothing they record or make goes to Lectern's servers. A shared project with questions keeps each viewer's answers for its owner, unless the owner turns that off; then the link sends nothing about who watches, which is how K–12 teachers share with students. Students in grades K–12 can't have seats in a school's plan. The privacy policy lists what a college student's account holds, what it doesn't, and who handles it.

A security analyst asks

Can we block its network access entirely?

Yes. Recording, editing and export don't notice; going live, sync, share links, updates and crash reports won't work. Captions need the speech model, which you can put in place yourself, and reading a photo of a page needs its language file. On Windows, choose Whisper in Settings so voice follow stays on the device.

An accessibility coordinator asks

Is the app itself accessible?

Its text and controls are held to WCAG AA contrast, and every change is checked with automated accessibility tests. A formal conformance report hasn't been written.

An administrator asks

Who do we talk to?

Write to hello@lecternvideo.com. For the rest of the product, start at the front page.

Ready when you are.

Lectern is at version 0.1.16, for macOS, Windows and Linux.

Get the app