Abstract: In this work, we introduce the first autoregressive framework for real-time, audio-driven portrait animation, a.k.a, talking head. Beyond the challenge of lengthy animation times, a critical ...
FastAPI service around Kokoro‑82M with a persistent WebSocket using an OpenAI Realtime‑compatible event schema for low‑latency sentence‑level streaming and barge‑in. Output is 24 kHz mono PCM16 ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback