The phone part is real.
Time to first audio:
- Pixel 9: under 800 ms
- M3 MacBook Air in browser: under 450 ms
Both generate speech in real time with ZONOS3-Mini running locally.
Trained on @NVIDIA H100s. Served on H200s on Zyphra Inference.
Zyphra reports ZONOS3-Mini speech generation on Pixel 9 and MacBook Air
AIZyphra says its ZONOS3-Mini speech model runs locally on a Pixel 9 phone, with time to first audio under 800 ms. On an M3 MacBook Air in the browser, time to first audio is under 450 ms, and both devices generate speech in real time.
@ZyphraAI
Part of a thread · earlier post