# NVIDIA/personaplex

> PersonaPlex is a full-duplex speech-to-speech model built on Moshi, controlled by text role prompts and audio voice conditioning, for low-latency conversations with a consistent persona.

- Magnitude: 4.9 out of 10 — Breakout
- Stars: 11,070 total · +6 stars measured 2026-10-13, 10:53–12:26 UTC, ≈ 38 by evening
- Star trust: star growth looks organic
- Category: Generative media · Language: Python · License: MIT · Created: 2026-01-05 · Last push: 2026-03-02
- GitHub: https://github.com/NVIDIA/personaplex · Page: https://gitnova.dev/en/r/NVIDIA/personaplex

## Useful for

- Launch the local server and test an assistant dialogue via the Web UI at localhost:8998
- Build an offline pipeline that turns an input wav into an output wav with a chosen voice
- Test a customer-service role prompt against a sample audio file

## Why it’s here

- Star-counter measurements on 2026-10-13 (UTC), 10:53–12:26: 11064 → 11070 stars (+6). This is the change over that interval.
- Estimated end-of-day forecast: about +38 stars, using observed gains and the previous day.
- Before this, the repository barely got any stars — about 3 per day.
- Elevated gains were measured on the last 2 completed UTC days in a row. Continued growth is not guaranteed.
- GitHub Trending Python today: #11, +123 stars.
- About 19 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 1,517
- Issues and pull requests: 107
- Watchers: 101
- Average over the last week: 70 per day
- Usual pace: 4 per day
- Stars in the last hour (measured): 5

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-14 … 2026-10-13: 2, 7, 4, 3, 2, 0, 9, 4, 9, 3, 7, 1, 4, 3, 5, 2, 4, 3, 1, 3, 3, 1, 3, 0, 2, 3, 8, 311, 131, 6

## Spotted in now

- GitHub Trending Python today: #11, +123 stars

## Similar by description

1. **speakrail/speakrail** — 1.2 · Cooling · AI agents · Python · +1 star measured 2026-10-13, 00:07–12:38 UTC, ≈ 4 by evening
   A full-duplex voice assistant running fully locally on a single RTX 4090: it listens while speaking, handles interruptions and pauses, and replies in 0.7–0.8 s.
   Full card: https://gitnova.dev/en/r/speakrail/speakrail.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-13 12:43 UTC, updated every 30 minutes.