NVIDIA releases NemotronLabs VoiceChat 11B: open-source full-duplex speech model with about 450 ms turn-taking and real-time tool calling
Original title:NVIDIA Releases NemotronLabs VoiceChat 11B: An Open Full-Duplex Speech-to-Speech Model with ~450 ms Turn-Taking and Live Tool Calling
AI summary
NVIDIA released NemotronLabs VoiceChat 11B, an open-source end-to-end full-duplex speech dialogue model that performs streaming speech understanding and generation in a unified network, with measured turn-taking latency of 448 ms. It is the first open-source full-duplex model to support tool calling during conversation, using a separate output channel and preset hold phrases to avoid silence during API execution. Weights and containers are public but for research use only, require a single 80 GB GPU, and no hosted API is currently available.
Key Takeaways
- NVIDIA released NemotronLabs VoiceChat 11B,an open-source end-to-end full-duplex speech dialogue…
- It is the first open-source full-duplex model to support tool calling during conversation,using a…
- Weights and containers are public but for research use only,require a single 80 GB GPU,and no…
Quick Facts
- Published
- Updated
- Source
- MarkTechPost(RSS)
- Category
- Model Releases
- AI score
- 75
端到端语音模型消除级联延迟,而工具调用侧通道和保留消息设计在不中断对话的前提下补齐了API等待期,让实时语音智能体更容易工程化。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: NVIDIA releases NemotronLabs VoiceChat 11B: open-source full-duplex speech model with about 450 ms turn-taking and real-time tool calling. August 10, 2026. https://new.nat6.net/en/news/cmsmhbu4w0338ronxh92vej2u/