Meta's speech model targets messy real rooms as the ears for its glasses and agent stack
heypearlai · x · 2026-09-04
Meta's new streaming speech system is positioned as the "ears" for its AI glasses and agent stack, built to handle messy real-world rooms with multiple speakers rather than waiting for a clean "hey Meta" wake command.
- It is live through the Meta Model API, Meta AI for Mac, and Muse Code.
- The author stays skeptical: current demos ran with 20 consenting recorded participants, and performance in an actual chaotic living room remains unproven.
More from Embodied
- Sentdex trains robots to go from lying down to a half kneel, eyeing a full stand — Sentdex · 2026-10-21
- NVIDIA open-sources MotionBricks: 350K motion skills at 15,000 FPS, 2ms latency — aftahi_ai · 2026-10-21
- NVIDIA Cosmos3-H-Dreams: Generative Simulation Brings Clinical Surgery to Life — Hugging Face Blog · 2026-10-21
- FLoTS-SLAM uses optical-flow residuals to hit 0.73 DM-IoU in dynamic monocular SLAM — zhenjun_zhao · 2026-10-21
- Plato ring launches at $199 to turn near-silent speech into AI commands — TinfoilTricorn · 2026-10-21
- Dream-1 controlled 31 unseen robot types across 53 countries within 24 hours — gharik · 2026-10-21