NVIDIA Releases First Open-Source Full-Duplex Voice Model

NVIDIA has released Nemotron VoiceChat-11B on Hugging Face, marking the first open-source full-duplex voice model. Based on Nemotron-Nano, it supports natural turn-taking, interruptions, and tool calling for voice agents.

2026-08-04 ~ 2026-08-04 · 3 related posts

• NVIDIA posts NemotronLabs VoiceChat 11B on Hugging Face as a full-duplex model — adefa · 2026-08-04
• NVIDIA Releases Nemotron VoiceChat: First Open Full-Duplex Speech Model with Tool Calling — _akhaliq · 2026-08-04
• NVIDIA Releases VoiceChat-11B Model on Hugging Face — nvidia · 2026-08-04