nvidia/NVIDIA-NemotronLabs-VoiceChat-11B · Hugging Face

Log In<br>Sign Up

","children":[],"isValid":true,"title":"Use Case: "},{"id":"deployment-geography","label":"Deployment Geography","children":[],"isValid":true,"title":"Deployment Geography"},{"id":"release-date--","label":"Release Date: ","children":[],"isValid":true,"title":"Release Date: "}],"isValid":true,"title":"Description:"},{"id":"model-architecture","label":"Model Architecture","children":[{"id":"input","label":"Input","children":[],"isValid":true,"title":"Input"},{"id":"output","label":"Output","children":[],"isValid":true,"title":"Output"}],"isValid":true,"title":"Model Architecture"},{"id":"how-nemotronlabs-voicechat-compares-to-other-open-fd-models","label":"How NemotronLabs VoiceChat Compares to Other Open FD Models","children":[],"isValid":true,"title":"How NemotronLabs VoiceChat Compares to Other Open FD Models"},{"id":"software-integration","label":"Software Integration","children":[],"isValid":true,"title":"Software Integration"},{"id":"model-versions","label":"Model Version(s)","children":[],"isValid":true,"title":"Model Version(s)"},{"id":"quickstart","label":"Quickstart","children":[],"isValid":true,"title":"Quickstart"},{"id":"training-testing-and-evaluation-datasets","label":"Training, Testing, and Evaluation Datasets:","children":[{"id":"training-dataset","label":"Training Dataset:","children":[],"isValid":true,"title":"Training Dataset:"},{"id":"testingevaluation-dataset","label":"Testing/Evaluation Dataset:","children":[],"isValid":true,"title":"Testing/Evaluation Dataset:"},{"id":"offline-inference","label":"Offline inference","children":[],"isValid":true,"title":"Offline inference"},{"id":"interactive-streaming-deployment","label":"Interactive streaming deployment","children":[],"isValid":true,"title":"Interactive streaming deployment"}],"isValid":true,"title":"Training, Testing, and Evaluation Datasets:"}],"isValid":true,"title":"Model Overview"},{"id":"inference","label":"Inference:","children":[{"id":"function-calling-system-prompt-example","label":"Function-calling system prompt example","children":[],"isValid":true,"title":"Function-calling system prompt example"},{"id":"known-limitations","label":"Known Limitations","children":[],"isValid":true,"title":"Known Limitations"},{"id":"ethical-considerations","label":"Ethical Considerations","children":[],"isValid":true,"title":"Ethical Considerations"},{"id":"references","label":"References:","children":[],"isValid":true,"title":"References:"}],"isValid":true,"title":"Inference:"}]}">

NVIDIA NemotronLabs VoiceChat 11B

▶ Hear it first. Natural turn-taking, barge-in and live tool calling.

Sample<br>What to listen for<br>Player

Natural turn-taking<br>Smooth back-and-forth, ~450 ms response

Barge-in / interruption<br>User cuts in; model yields instantly

Tool calling live<br>Tools are called when needed

▶ Code: On Github

Model Overview

Description:

NVIDIA NemotronLabs VoiceChat is a 11B end-to-end, real-time speech full duplex (FD) model for conversational AI that jointly performs streaming speech understanding and speech generation [1, 2]. Unlike traditional cascaded stacks (ASR → LLM → TTS), this model achieves full duplex, real-time, seamless voice interaction in one unified architecture, eliminating the need for multiple models or API handoffs, thus reducing end-to-end latency.<br>It sets new benchmarks by bringing open, robust, and highly natural conversation capabilities. Moreover, NVIDIA NemotronLabs VoiceChat is the first open full-duplex model to support tool calling while maintaining a natural conversation flow during tool execution. For each tool, a specific “on-hold” message can be defined that will be spoken by the agent as soon as the LLM generates the text that will trigger the tool call and response.

The model operates on audio signals, which are encoded using a fast conformer module. The resulting audio tokens are inputted into a Nemotron Nano V2 9B LLM backbone to predict text tokens, which are fed to a TTS decoder [2] to predict audio codes for generating the agent's speech. A separate output channel is used to predict tool calling scripts.<br>NemotronLabs VoiceChat offers an unprecedented trade-off between intelligence and latency in the space of open-source voice agents, as highlighted by our benchmarking results below.

This model is ready for research purposes only.

Highlights

Parameters<br>Turn-taking latency<br>VoiceBench (open FD)<br>Tool calling<br>Architecture

11B<br>~450 ms<br>#2<br>1st open FD<br>1 unified model

License/Terms of Use

GOVERNING TERMS: The use of this model is governed by the OpenMDW License Agreement, version 1.1.

Use Case:

NemotronLabs VoiceChat is targeted for researchers, developers, and professionals in the field of natural language processing (NLP) and speech technology for purposes such as automatic speech recognition (ASR), text-to-speech synthesis (TTS), and voice assistant development.

Deployment Geography

Global

Release Date:

August 3,...