Tether launches offline AI speech model for smart glasses and robots
- —254M-parameter model transcribes 10 minutes of audio in 1.65 seconds on iPhone 17 Pro Max
- —Supports 25 European languages on laptops and eight on phones
- —Word error rate of 8.02% on Open ASR Leaderboard and 6.58% on Multilingual LibriSpeech
- —Release is part of Tether's open-source QVAC AI project for on-device AI
Why it matters: Tether is expanding beyond stablecoins into decentralized AI, strengthening its position in private on-device computing.
Source: crypto.news