Traditional captioning pipelines upload your video to a server, transcribe it in the cloud, render it, and send a link back. Every step is a copy of your content sitting on someone else's infrastructure.
SnipCaptions flips that model. The browser extracts audio, sends only a short transcription request to Gemini using your key, and renders the final MP4 locally with WebCodecs. The source file is never uploaded to us.
Cost and speed
With no server GPUs to pay for, the product can stay free for HD exports. And because rendering is local and hardware-accelerated, there is no queue to wait in.