Subvox: one-prompt, 9-minute rebuild of silent-speech recognition using AirPods Pro motion data
seatedro · x · 2026-10-31
joeblau recreated "subvocal" in 9 minutes with a single prompt, open-sourced as Subvox — a macOS prototype testing whether AirPods Pro head-motion data (CMHeadphoneMotionManager) carries enough signal to turn silently mouthed words into text. No language model: just a small-vocabulary template recognizer plus a statistical test. Repo includes full xcodegen/Xcode build steps (macOS 14+). As seatedro quipped: it's always "recreated," never "created" — AI now makes recreating demos nearly free.
More from Embodied
- China's dark warehouses run robots 24/7 — but do humanoids really fit? — luisdans · 2026-10-31
- Xiaomi's fabric-covered robotic hand shows off a coin-rolling finger trick — luisdans · 2026-10-31
- Will robots be solved by LLMs writing control code in real time? Debate sparks on X — lukas_m_ziegler · 2026-10-31
- Hy-Nomas: open-source interactive world model runs on consumer GPUs, rivals Genie 4 — bdsqlsz · 2026-10-31
- Tesla catches Waymo's Texas robotaxi fleet size in just 4 months vs 18 — mitchdeg · 2026-10-31
- GOFish: fisheye-trained 3D Gaussian Splatting surpasses 180° FoV and extracts collision-ready meshes — janusch_patas · 2026-10-31