SubExtractor

What we don't collect

SubExtractor is designed privacy-first. We do not collect, store, or have access to:

• Personal information (name, email, phone, address)
• Video content, audio, captions or transcripts
• Browsing history or the sites you visit
• Device information, hardware identifiers, or fingerprints
• Location data
• Analytics or usage-tracking data
• IP addresses for tracking

What is processed locally

All processing happens on your device. The following never leaves it:

• Caption text extracted from videos
• Audio from a browser tab you choose to transcribe (Chrome), and the resulting transcript
• Captions read aloud — the voice is generated on your device (Chrome)
• Your preferences, settings and recent results
• Downloaded speech-recognition language packs and voices (see Downloads below), kept on your device until you remove them or uninstall

Platform data

To extract captions, SubExtractor reads captions from the video platform you're viewing (e.g. YouTube) using that platform's own systems. Only the minimal caption metadata needed to display results is used, and it is processed on your device. No platform identifiers, accounts, or logins are tracked or stored.

Downloads (Chrome extension)

Some features need files the first time you use them. These downloads contain only models and voices — no audio, transcript or page data is sent with them — but, as with any download, the provider can see your IP address.

• Live transcription downloads a speech-recognition language pack from Moonshine AI (download.moonshine.ai).
• Optional speaker labels download speaker-identification models from the same host.
• Read captions aloud downloads a voice from the same host. You can remove it from the extension at any time.
• Translation, summaries and clean-up use Chrome's own on-device AI; Chrome downloads and manages those models itself.
• The extension loads its font from Google Fonts.

Browser permissions (Chrome & Safari extension)

The extension requests only the permissions its features need:

• Site access — to read caption data on the video page you're viewing, and to show captions over it. Captions are processed locally and never transmitted to us.
• Active tab — to act on the tab you're using.
• Tab capture (Chrome) — to transcribe the audio of a tab, only when you start it. The audio is processed on your device and never recorded to a server.
• Web request (Chrome) — to notice subtitle files a video page loads, so they can be extracted. Their addresses are kept on your device only.
• Storage and unlimited storage — to save your settings on your device, and to keep downloaded language packs and voices so they are not downloaded again.
• Offscreen, scripting and side panel (Chrome) — to run speech recognition in the background, show captions on the page, and display your transcript.

To retrieve captions, the extension fetches the caption or subtitle file directly from the video site you're viewing (for example YouTube, Vimeo, or TED) using that site's own systems. That request goes to the video platform you already have open — never to us. No captions, transcripts, audio, or browsing data are ever sent to SubExtractor's developer or to any third-party server.

Third-party services

SubExtractor uses no third-party analytics, advertising, or tracking services. The only external requests are caption files from the video site you have open and the downloads described above — none of them carry your audio, transcripts, or browsing data.

Children's privacy

SubExtractor collects no personal data from anyone, including children under 13.

Changes to this policy

If this policy changes, the updated date above will change. Material changes will be reflected in the app or extension listing.

---

Contact

Questions about this privacy policy? Email [email protected].