What do you want to do?
3 use cases
-
Talk to a full-duplex voice model in your browser
Start the local server and hold a real-time spoken conversation through the web UI.
- REPO
NVIDIA/personaplex - Research
- Advanced
View prompt
Install NVIDIA/personaplex, start moshi.server with SSL, and hold a 1 minute conversation. Report latency observations and confirm the server ran on GPU.
- REPO
-
Give a voice agent a role and a voice
Combine a text role prompt with a voice prompt to keep a consistent persona.
- REPO
NVIDIA/personaplex - Research
- Advanced
View prompt
Run PersonaPlex offline with voice prompt <file>, role prompt <text> and input <wav>, then check output.json follows the role.
- REPO
-
Generate spoken replies from recorded input offline
Run offline inference on a WAV file to get the model's spoken reply and transcript.
- REPO
NVIDIA/personaplex - Research
- Advanced
View prompt
Run moshi.offline from NVIDIA/personaplex on <input.wav> with seed 42424242 and two different voice prompts. Report differences between the two outputs.
- REPO
Quick start with Personaplex
- Check you have what it needs: Linux with an NVIDIA GPU; Opus development library; A Hugging Face token with access to the model.
- Open NVIDIA/personaplex on GitHub and read the README and license.
- Install Opus with
sudo apt install libopus-dev - Install with
pip install moshi/.and setexport HF_TOKEN= - Start with
SSL_DIR=$(mktemp -d); python -m moshi.server --ssl "$SSL_DIR" - Open the web UI address the server prints
These steps come from our top use case, Talk to a full-duplex voice model in your browser, which also has a copy-paste agent prompt.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for NVIDIA/personaplex
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeAbout this repo
- Who it is forSpeech AI researchers and developers building low-latency voice agents with a consistent persona.
- Why it is hot+270 stars this week, 11k total, picked up by one AI source we track
- LanguagePython
- Sourcegithub.com/NVIDIA/personaplex · Back to Hot repos
Requirements
- Linux with an NVIDIA GPU
- Opus development library
- A Hugging Face token with access to the model
Add this badge to your README
Maintain Personaplex? Show readers where to find these use cases.
Markdown
[](https://dedupedai.com/use-cases/nvidia-personaplex/?utm_source=github&utm_medium=badge&utm_campaign=NVIDIA-personaplex)
HTML
<a href="https://dedupedai.com/use-cases/nvidia-personaplex/?utm_source=github&utm_medium=badge&utm_campaign=NVIDIA-personaplex"><img src="https://dedupedai.com/badge/NVIDIA/personaplex.svg" alt="Use cases on Deduped" height="20"></a>
A static image, no tracking script. About the badge
Questions about Personaplex
What is Personaplex?
PersonaPlex is NVIDIA's real-time, full-duplex speech-to-speech conversational model based on Moshi that controls persona through text role prompts and audio voice conditioning. It is an open-source project on GitHub: https://github.com/NVIDIA/personaplex.
Who is Personaplex for?
Speech AI researchers and developers building low-latency voice agents with a consistent persona.
What can I do with Personaplex?
3 use cases so far, for example: Talk to a full-duplex voice model in your browser; Give a voice agent a role and a voice; Generate spoken replies from recorded input offline. Each has steps and a copy-paste agent prompt.
What do I need to run Personaplex?
Linux with an NVIDIA GPU; Opus development library; A Hugging Face token with access to the model.
What are alternatives to Personaplex?
Other hot AI repos in Research: Scientific Agent Skills, Trustworthy Agent Simulation, Kaas, By2 KB, Lecture to Notes.
Is Personaplex safe to run?
Deduped does not own, audit or vouch for it. Read the README and license, run it on your own systems first and review every command before you run it.