personaplex README

The project's README from github.com/NVIDIA/personaplex. As one markdown file: README.md.

PersonaPlex: Voice and Role Control for Full Duplex Conversational Speech Models

Weights Paper Demo Discord

PersonaPlex is a real-time, full-duplex speech-to-speech conversational model that enables persona control through text-based role prompts and audio-based voice conditioning. Trained on a combination of synthetic and real conversations, it produces natural, low-latency spoken interactions with a consistent persona. PersonaPlex is based on the Moshi architecture and weights.

PersonaPlex Model Architecture
PersonaPlex Architecture

Usage

Prerequisites

Install the Opus audio codec development library:

# Ubuntu/Debian
sudo apt install libopus-dev

# Fedora/RHEL
sudo dnf install opus-devel

Installation

Download this repository and install with:

pip install moshi/.

Extra step for Blackwell based GPUs as suggested in (See https://github.com/NVIDIA/personaplex/issues/2):

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu130

Accept Model License

Log in to your Huggingface account and accept the PersonaPlex model license here.
Then set up your Huggingface authentication:

export HF_TOKEN=<YOUR_HUGGINGFACE_TOKEN>

Launch Server

Launch server for live interaction (temporary SSL certs for https):

SSL_DIR=$(mktemp -d); python -m moshi.server --ssl "$SSL_DIR"

CPU Offload: If your GPU has insufficient memory, use the --cpu-offload flag to offload model layers to CPU. This requires the accelerate package (pip install accelerate):

SSL_DIR=$(mktemp -d); python -m moshi.server --ssl "$SSL_DIR" --cpu-offload

Access the Web UI from a browser at localhost:8998 if running locally, otherwise look for the access link printed by the script:

Access the Web UI directly at https://11.54.401.33:8998

Offline Evaluation

For offline evaluation use the offline script that streams in an input wav file and produces an output wav file from the captured output stream. The output file will be the same duration as the input file.

Add --cpu-offload to any command below if your GPU has insufficient memory (requires accelerate package). Or install cpu-only PyTorch for offline evaluation on pure CPU.

Assistant example:

HF_TOKEN=<TOKEN> \
python -m moshi.offline \
  --voice-prompt "NATF2.pt" \
  --input-wav "assets/test/input_assistant.wav" \
  --seed 42424242 \
  --output-wav "output.wav" \
  --output-text "output.json"

Service example:

HF_TOKEN=<TOKEN> \
python -m moshi.offline \
  --voice-prompt "NATM1.pt" \
  --text-prompt "$(cat assets/test/prompt_service.txt)" \
  --input-wav "assets/test/input_service.wav" \
  --seed 42424242 \
  --output-wav "output.wav" \
  --output-text "output.json"

Voices

PersonaPlex supports a wide range of voices; we pre-package embeddings for voices that sound more natural and conversational (NAT) and others that are more varied (VAR). The fixed set of voices are labeled:

Natural(female): NATF0, NATF1, NATF2, NATF3
Natural(male):   NATM0, NATM1, NATM2, NATM3
Variety(female): VARF0, VARF1, VARF2, VARF3, VARF4
Variety(male):   VARM0, VARM1, VARM2, VARM3, VARM4

Prompting Guide

The model is trained on synthetic conversations for a fixed assistant role and varying customer service roles.

Assistant Role

The assistant role has the prompt:

You are a wise and friendly teacher. Answer questions or provide advice in a clear and engaging way.

Use this prompt for the QA assistant focused "User Interruption" evaluation category in FullDuplexBench.

Customer Service Roles

The customer service roles support a variety of prompts. Here are some examples for prompting style reference:

You work for CitySan Services which is a waste management and your name is Ayelen Lucero. Information: Verify customer name Omar Torres. Current schedule: every other week. Upcoming pickup: April 12th. Compost bin service available for $8/month add-on.

You work for Jerusalem Shakshuka which is a restaurant and your name is Owen Foster. Information: There are two shakshuka options: Classic (poached eggs, $9.50) and Spicy (scrambled eggs with jalapenos, $10.25). Sides include warm pita ($2.50) and Israeli salad ($3). No combo offers. Available for drive-through until 9 PM.

You work for AeroRentals Pro which is a drone rental company and your name is Tomaz Novak. Information: AeroRentals Pro has the following availability: PhoenixDrone X ($65/4 hours, $110/8 hours), and the premium SpectraDrone 9 ($95/4 hours, $160/8 hours). Deposit required: $150 for standard models, $300 for premium.

Casual Conversations

The model is also trained on real conversations from the Fisher English Corpus with LLM-labeled prompts for open-ended conversations. Here are some example prompts for casual conversations:

You enjoy having a good conversation.

You enjoy having a good conversation. Have a casual discussion about eating at home versus dining out.

You enjoy having a good conversation. Have an empathetic discussion about the meaning of family amid uncertainty.

You enjoy having a good conversation. Have a reflective conversation about career changes and feeling of home. You have lived in California for 21 years and consider San Francisco your home. You work as a teacher and have traveled a lot. You dislike meetings.

You enjoy having a good conversation. Have a casual conversation about favorite foods and cooking experiences. You are David Green, a former baker now living in Boston. You enjoy cooking diverse international dishes and appreciate many ethnic restaurants.

Use the prompt You enjoy having a good conversation. for the "Pause Handling", "Backchannel" and "Smooth Turn Taking" evaluation categories of FullDuplexBench.

Generalization

Personaplex finetunes Moshi and benefits from the generalization capabilities of the underlying Helium LLM. Thanks to the broad training corpus of the backbone, we find that the model will respond plausibly to out-of-distribution prompts and lead to unexpected or fun conversations. We encourage experimentation with different prompts to test the model's emergent ability to handle scenarios outside its training distribution. As an inspiration we feature the following astronaut prompt in the WebUI:

You enjoy having a good conversation. Have a technical discussion about fixing a reactor core on a spaceship to Mars. You are an astronaut on a Mars mission. Your name is Alex. You are already dealing with a reactor core meltdown on a Mars mission. Several ship systems are failing, and continued instability will lead to catastrophic failure. You explain what is happening and you urgently ask for help thinking through how to stabilize the reactor.

License

The present code is provided under the MIT license. The weights for the models are released under the NVIDIA Open Model license.

Citation

If you use PersonaPlex in your research, please cite our paper:

@misc{roy2026personaplexvoicerolecontrol,
      title={PersonaPlex: Voice and Role Control for Full Duplex Conversational Speech Models}, 
      author={Rajarshi Roy and Jonathan Raiman and Sang-gil Lee and Teodor-Dumitru Ene and Robert Kirby and Sungwon Kim and Jaehyeon Kim and Bryan Catanzaro},
      year={2026},
      eprint={2602.06053},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2602.06053}, 
}

personaplex (NVIDIA/personaplex)

Searching Database…

11.2K ·GitHub
Personaplex
What is this?PersonaPlex is a real-time, full-duplex speech-to-speech conversational AI model from NVIDIA. It lets you control the assistant's persona through text role prompts and condition its voice on an audio sample, producing natural low-latency spoken conversations.
Is this a model I load in code, or does it come with a ready-to-use app?It ships as a server you launch locally, which exposes a Web UI you talk to from a browser. It's built on the Moshi architecture and finetuned Helium LLM backbone, not a standalone library.
How do I get it running?Install the Opus audio codec dev library, download and install the repo, accept the PersonaPlex model license on Hugging Face and set up Hugging Face authentication, then launch the server and open localhost:8998 in your browser.
Do I need a Hugging Face account and a specific GPU setup?Yes, you need a Hugging Face account to accept the model license and authenticate, plus the Opus codec library. Blackwell-based GPUs need an extra install step, and low-VRAM GPUs can use the --cpu-offload flag (requires the accelerate package).
What if my GPU doesn't have enough memory?You can pass --cpu-offload to move model layers to CPU, which requires installing the accelerate package; offline evaluation can also run on pure CPU with a CPU-only PyTorch build.
Who is this aimed at?It's aimed at researchers and developers experimenting with full-duplex conversational speech models, persona/voice control, and real-time speech interaction research.
What voices are available out of the box?PersonaPlex pre-packages a fixed set of voice embeddings labeled NAT (natural, conversational-sounding) and VAR (more varied), so you can assign a voice without recording your own sample.
How does persona control actually work?You give the model a text role prompt (e.g. an assistant or customer-service persona) combined with an audio voice sample, and it stays in that persona and voice throughout a live spoken conversation.
Can I test it without a live microphone session?Yes, there's an offline evaluation script that streams in an input WAV file and produces an output WAV file of the same duration, useful for benchmarking responses to prewritten prompts.
What is Moshi, and why does it matter here?Moshi is the underlying full-duplex speech architecture and weight set PersonaPlex is finetuned from; it also inherits generalization ability from the Helium LLM backbone, letting it respond plausibly even to out-of-distribution prompts.
What are similar projects?Retrieval-based-Voice-Conversion-WebUI, VoxCPM, csm, fish-speech, Index-Translate, FireRedAudio, Real-Time-Voice-Cloning, moshi
Which topics is it in?speech, conversational-ai, voice-cloning, real-time, audio, inference, nvidia, huggingface, cuda, pytorch, deep-learning, research
Creator
NVIDIA
Size
1.3 MB
Stars
11.2K
Updated
CommitMar 02
Age
9mo
Commits · 90 Days