Updates

Search now covers everything you've processed. Open the command menu (⌘K / Ctrl+K) and search titles and transcript content across public videos, private uploads, and imported playlists; a result takes you straight into the matching video.

Transcript downloads

The transcript copy menu now downloads files: plain text (TXT) for documents, and SRT subtitles ready for YouTube or a video editor, alongside the clipboard copies and CSV export. SRT timestamps come from the transcript segments themselves, so they line up with the player.

Your library is now accessible from outside Video Highlight.

Personal access tokens

Create and revoke tokens in Settings, then API Keys. Tokens are shown once at creation, work with the public API, and revocation takes effect immediately.

MCP server (beta)

The MCP server exposes your library to any assistant that speaks the Model Context Protocol: list your videos, fetch transcripts and summaries, and search across everything you've processed, from inside the assistant.

API access is available on Pro and Business plans and is marked experimental; endpoints may change. If you build something with it, write to support@videohighlight.com. Early feedback shapes what's exposed next.

The app now shares one visual language from the video pages to Settings, and every screen reads the same way, including on small displays.

Improvements

• A borderless header shell with a clear page title on every screen.
• One page-header system for the video AI pages and all of Settings.
• Pill-shaped form controls and a rounder component style throughout.
• Empty states render in place while data loads.
• A single brand wordmark shared between the app and the website.

Leaving should be as easy as joining. You can now cancel or change your subscription from Settings, then Subscriptions, with exactly what happens to your remaining time shown before you confirm.

Cancelling keeps your plan active until the end of the paid period. Your library, summaries, and transcripts stay intact on the Free plan afterwards. If something prompted the change, the flow lets you say so in a sentence.

Speech recognition is most accurate when it knows what it's listening to. You can now choose the spoken language when you upload a private file, and recognition uses the best model for that language from the first second. Accuracy holds up on noisy audio, strong accents, regional dialects, and short clips, and names, numbers, and punctuation come through intact.

Open the picture at full size

Auto-detect remains available. The choice affects speech recognition only; the summary language is set separately. If you mostly work in one language, set a default in Preferences.

Summaries and transcriptions now cover 37 languages. The language selector shows an accuracy indicator for each one ( High, Medium, Low ).

Open the picture at full size

High Accuracy

• English
• Spanish
• Portuguese
• Italian
• German
• Russian
• Polish
• French
• Turkish
• Indonesian
• Dutch
• Malay
• Ukrainian
• Korean

Medium Accuracy

• Chinese
• Swedish
• Vietnamese
• Thai
• Finnish
• Norwegian
• Filipino
• Romanian
• Arabic
• Greek
• Danish
• Slovak
• Croatian

Low Accuracy

• Urdu
• Bulgarian
• Czech
• Hindi
• Hungarian
• Hebrew
• Estonian
• Latvian
• Welsh
• Japanese

To request a language or regional variant, write to support@videohighlight.com.

Research is rarely a solo activity. Workspaces bring your team into Video Highlight: every account starts with a personal workspace, and on paid plans you can invite teammates and share playlists, transcripts, and highlights with them.

Members and roles

Invite collaborators by email or invite link, as Admin or Member. Transcripts, highlights, and playlists are shared with the workspace, so everyone works from the same material.

Switching

Move between workspaces from the user menu or the Settings sidebar. Your personal workspace is created on sign-up and can be renamed at any time.

Workspaces

Documentation: Workspaces overview · Members and roles · Switching and sharing