voice-ai

Build Voice AI with nemotron-voice-agent

Name: Build Voice AI with nemotron-voice-agent
Author: NVIDIA-AI-Blueprints

Get a reference voice agent implementation for your startup, built with NVIDIA Nemotron models, for founders needing voice AI.

From the repo · “Reference implementation of an end-to-end voice agent built using the NVIDIA Nemotron models”

246 stars+5★ this week68 forksTypeScriptGuide quality 9/10 · reviewed Jul 2Updated 10/21/2026100% free · open source

What it does

This tool provides a reference implementation of an end-to-end voice agent built using the NVIDIA Nemotron models, allowing founders to integrate voice AI into their startup

When to use it

•When you need to build a conversational interface for your application
•When you want to leverage pre-trained voice AI models for faster development
•When you're looking for a customizable voice agent solution for your specific use case

Ready-to-paste prompt

node inference.js --audio_file examples/audio_sample.wav --model_path models/nemotron.tlt

Heads up: Ensure you have the necessary NVIDIA dependencies installed, including the NVIDIA TensorRT and cuDNN libraries, to run the Nemotron models

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🔑 Needs an API key

Works after you add credentials — the setup agent will ask you for them.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/nemotron-voice-agent && curl -fsSL https://workflowstacks.com/api/skills/nemotron-voice-agent/claude-skill -o ~/.claude/skills/nemotron-voice-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Build Voice AI with nemotron-voice-agent works

Codeflow

Free to inspect

Build Voice AI with nemotron-voice-agent is a large Python project (~35k lines across 167 code files, plus 52 test files). Setup is light: runs with Docker — add your API keys to .env. Last commit this month, Other license, has a test suite.

Size

Large codebase

~35k lines · 167 code files · ~5 h to skim

Setup

Light setup

Runs with Docker — add your API keys to .env.

Runs on

Python · Docker

Needs API keys (.env)

Python 83%TypeScript 11%SCSS 2%Shell 1%Jupyter Notebook 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
src/server.py
Where the program starts running
4
pyproject.toml
Dependencies and the commands it exposes
5
.env.example
The API keys and settings you must provide

What's in each folder

src/Core code — the actual logic119 files

skills/Prompts, skills & agent definitions29 files

client/Frontend / UI54 files

docs/Documentation22 files

benchmarking_tools/Folder22 files

docker/Deployment / infrastructure14 files

scripts/Helper scripts2 files

models/Data models & types1 files

READMEHas testsDocumentedCI checksDocker readyOther licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🔑 Needs an API key

Works after you add credentials — the setup agent will ask you for them.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/nemotron-voice-agent && curl -fsSL https://workflowstacks.com/api/skills/nemotron-voice-agent/claude-skill -o ~/.claude/skills/nemotron-voice-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

NVIDIA-AI-Blueprints

Language

TypeScript