Tencent open-sources Hy-SimulMT, a ~3s-latency real-time CN↔EN interpreting model with voice cloning
TencentHunyuan · x · 2026-10-26
Tencent Hunyuan has open-sourced Hy-SimulMT, a real-time Chinese↔English simultaneous interpreting model, available on Hugging Face, ModelScope and GitHub.
- 3s latency, aimed at live interpreting scenarios
- Real-time voice cloning to preserve speaker timbre
- Main checkpoint is Hy-SimulMT-30B-A3B, shipped in FP8, GGUF and Interleave variants
- The team claims it outperforms most open-source models and commercial APIs
Developers can build real-time translation and live-captioning tools on top of it.
More from Multimodal
- Claude Code can now edit videos: a 10-prompt workflow from raw footage to final cut — TawohAwa · 2026-10-26
- One-sentence 3D worlds turn cinematic with CineMap and TapNow — gizakdag · 2026-10-26
- No Camera Needed: Creator Builds an Entire World in CineMap and Directs It Into Cinema — gizakdag · 2026-10-26
- Kling vs. Seedance: the Chinese AI video model rivalry — DavidmComfort · 2026-10-26
- Bokeh Cosmic Nebula prompt template: out-of-focus orbs mimic interstellar clouds — LudovicCreator · 2026-10-26
- AIandDesign reveals Runway Act-2 + ElevenLabs voice changer video workflow — AIandDesign · 2026-10-26