jevVIDEO

Jev routes on-device models to turn voice memos into to-dos

In a demo app, Ear detects the language, Voz transcribes and Redact strips PII, while Jev makes about 20 decisions in one call within milliseconds to pick which on-device model to run, with no LLM in the loop.

Desert Ant Labs

@desertantlabsArchived 2026.09.23Originally posted 2026.09.22

Desert Ant Labs @desertantlabsOriginally posted 2026.09.22

Open original on X

Jev + on-device models = results in seconds with no LLM in the loop. Quick demo app to show the possibilities. Drop in an audio file: Ear detects the language, Voz transcribes it and Redact removes PII. Then @typesafeai's Jev makes about 20 decisions in one call in milliseconds, and picks which of our on-device models to run. Voice memo to to-do list. Meeting to redacted transcript. Podcast to clips.

x.com/desertantlabs/status/2102423007494889785Archived 2026.09.23 10:16