Home Server

Local speech-to-text for every recording.

Whisper runs on your home server's GPU or CPU. Every voice memo, every recorded conversation, every audio post — transcribed locally, indexed into your brain, searchable forever. Free, accurate, offline.

Choose Your Plan Back to overview

Local speech-to-text via Whisper — every recording transcribed on your home server, indexed and searchable.

What it does

Whisper STT at a glance.

Model picker

Tiny (fast, OK), Base (default), Small, Medium, Large-v3. Pick once in the Whisper app; trades speed for accuracy on your hardware.

Auto-transcribe

Voice memos, Live Studio recordings, family calls — anything captured to a brain folder auto-transcribes through a folder agent. No manual step.

Speaker diarization

Multi-participant conversations get per-speaker attribution. The Conversation Engine pairs face recognition with diarized turns to build accurate transcripts.

Search anywhere

OB1 reads transcripts the same as any other text. 'What did Sarah say about the gallery on the call last week?' just works.

Add Whisper STT to your agent.

Every VP3 plan unlocks the full c:// drive catalog. Pick a plan, install the home server, flip the app on.

Choose Your Plan