FAQ
Frequently Asked Questions
Everything you need to know about VideoChapter — how it works, what it supports, and what to expect.
General
What is VideoChapter?
VideoChapter is a native macOS app that turns a local video into chapters, summaries, a searchable transcript, grounded assistant answers, and exports — with core processing designed to stay on your Mac.
Is VideoChapter free?
Yes. The Free tier includes chapters, summaries, transcript search, and click-to-seek playback. Pro features such as assistant, translation, and exports unlock with a one-time purchase.
Does it really work offline?
Core processing runs locally after the one-time model download. Network access may still be used for downloads, updates, purchase checks, or optional anonymous analytics if you leave that enabled.
Technical
How large is the AI model download?
Expect roughly 8 GB of model files for the initial setup. The download happens once per machine unless you remove the models and install them again later.
What video formats are supported?
VideoChapter is designed around common formats that macOS media frameworks can open reliably, including MP4, MOV, and M4V. Unusual codecs or damaged files may not process correctly.
What Mac do I need?
macOS 26 or later is required. Apple Silicon is recommended for the best experience.
How long does processing take?
It depends on your Mac, the video length, the selected processing mode, and the audio quality. Shorter videos are practical to process locally; longer videos naturally take longer and use more memory.
What languages are supported for transcription and translation?
Transcription coverage depends on the bundled speech models. Translation depends on Apple-supported languages and any translation resources installed on your Mac.
Privacy & Security
Does my video get uploaded anywhere?
Not for the core processing workflow. Source videos, transcripts, chapters, summaries, and assistant answers stay local for the app’s core features.
Does the app ever contact external services?
Yes, when needed for model downloads, purchase or license flows, update checks, App Store restore flows, or optional anonymous analytics. Those flows are separate from core video processing.
Do you collect any data?
Core video processing does not upload your video or transcript content. Some builds can offer optional anonymous analytics, which can be disabled in Settings and should avoid sending transcript or video payloads.
Known Limitations
How reliable are speaker labels?
Speaker labeling can help separate voices, but overlapping speech, poor audio, and rapid turn-taking can still produce mistakes. Review speaker labels before publishing or relying on them.
Are there practical length limits?
Very long videos are supported, but processing time and memory use grow with duration. Large files can still be worth running locally if you expect a longer wait.
Can the assistant be wrong?
Yes. The assistant is grounded in transcript evidence and timestamps, which makes answers easier to verify, but you should still review outputs before using them in important contexts.
Have more questions?
Download VideoChapter and try the core workflow yourself before deciding if you need Pro.