Japanese Startup Open-Sources Full-Duplex Speech Model Haru-4B
Japanese startup higurashi AI has open-sourced Haru-4B, a full-duplex speech dialogue model topping Hugging Face trending charts. With 0.1s latency, barge-in handling, and support for 9 languages, it is being called a free open-source alternative to ChatGPT's voice mode.
2026-11-20 ~ 2026-11-20 · 3 related posts
- Haru-4B: open-source full-duplex voice model with 133ms frames, 9 languages — alex_verem · 2026-11-20
- Haru GitHub repo: low-latency multilingual full-duplex spoken dialogue model — alex_verem · 2026-11-20
- Japanese startup open-sources Haru-4B, a full-duplex voice AI that tops HF trending — solyarisoftware · 2026-11-20