Local-first captions
There is no managed or cloud caption service. Speech recognition and translation run on your own GPU, your audio never leaves the device, and there is no per-minute cloud bill.
Local-first captions
The public page should be clear that there is no cloud or managed caption mode: the AI runs on your GPU and your audio stays local.
There is no cloud caption service and no managed mode to switch on. Speech recognition and translation run on your own GPU, so your audio never leaves the device.
Because captions are generated locally, a flaky connection does not interrupt the show. Only account, licensing, and updates ever reach our servers.
Studios and multi-creator setups run several machines on one account, each captioning locally — Insider covers two devices and Sponsor three.
Choose the next product action that matches your workflow.