Support.

Get started on Mac

Approve the speech model download on first launch and wait for it to finish. Then choose Import MOV or MP4 in the sidebar or File > Import Video (Command-I), or open a MOV or MP4 with Silo Caption from Finder. Wait for recognition to finish; Cancel import is available while it runs. The spoken language for new imports is set in Settings, where auto-detect is also an option. Initial downloads need an internet connection; transcription processes the video on your Mac.

Recording and meeting notes

Silo Caption does not record audio or your screen, and it asks for no microphone, camera or Screen Recording access. It captions video files you import. To record and transcribe a lecture or interview, use Silo Transcribe; for meeting notes, use Silo Meet.

Edit, save and preview

Review the caption text, cue times and style. Click Save before Preview, subtitle export or video export. Unsaved edits disable those outputs. Review automatically generated captions before sharing them.

Karaoke highlights use the original recognized word timings only when the caption still matches those words. Edited, translated or unmatched cues remain static by design. Reduce the font size if the caption does not fit the chosen video shape.

Translate saved captions

Under Translate saved captions, choose a supported source and target language. Approve Apple's language download if prompted, then translate the saved revision. Select Original or a saved translation under Output before previewing or exporting. Editing the source captions can invalidate a saved translation; save the changes and translate again.

Language-pair availability depends on macOS. Offline translation requires the relevant language assets to be installed. Apple's framework has its own usage and performance metrics disclosure, explained in the privacy policy.

Export a new video or subtitle file

Choose MP4 or MOV for a captioned video, or SRT or VTT for a subtitle file. Video export offers landscape, portrait and square shapes with Standard or High resolution. Standard uses a 1080-pixel short edge; High uses a 2160-pixel short edge. High square output is 2160 × 2160.

Click Export and choose a new filename in the macOS save panel. Existing destination files are not replaced. Video export re-encodes source audio as AAC. Wait for the final completion message; Cancel is available while export is running.

If something goes wrong

• Model download failed: check your connection and available storage, then choose Try again in setup, or download the model again from model settings.
• No captions: confirm that the video contains audible speech and that the required speech model is installed. Recognition results need review.
• Preview or export is disabled: save the current caption edits first, then select the output track you want to use.
• Export failed: choose an unused destination with enough free space. Reduce text size if captions do not fit. Some unsupported media formats or timelines are refused instead of being silently altered.
• Launch error: note the exact message and contact support below. Do not delete your projects or model data to troubleshoot unless you have a backup and understand what will be removed.