---
title: VibeVoice
emoji: 🌍
colorFrom: purple
colorTo: yellow
sdk: docker
pinned: false
license: mit
short_description: VibeVoice-Realtime-0.5B - Real-time neural voice generation
---

<div align="center">

## �️ VibeVoice: Open-Source Frontier Voice AI
[![Project Page](https://img.shields.io/badge/Project-Page-blue?logo=microsoft)](https://microsoft.github.io/VibeVoice)
[![Hugging Face](https://img.shields.io/badge/HuggingFace-Collection-orange?logo=huggingface)](https://huggingface.co/collections/microsoft/vibevoice-68a2ef24a875c44be47b034f)
[![Technical Report](https://img.shields.io/badge/Technical-Report-red?logo=adobeacrobatreader)](https://arxiv.org/pdf/2508.19205)

**Real-time neural voice synthesis system powered by Microsoft VibeVoice-Realtime-0.5B.**  
*This Space demonstrates how to run the model using Docker on Hugging Face Spaces.*

</div>

<div align="center">
<picture>
  <source media="(prefers-color-scheme: dark)" srcset="Figures/VibeVoice_logo_white.png">
  <img src="Figures/VibeVoice_logo.png" alt="VibeVoice Logo" width="300">
</picture>
</div>


## 🚀 Space Features

- ⚡ **Real-time voice generation** (~300ms latency)
- 🧠 **Lightweight 0.5B parameter model**
- 🐳 **Docker-based deployment** (downloaded at runtime)
- 🌐 **Runs on CPU** (Zero GPU supported)

---

## 📦 Model Details

- **Model:** [`microsoft/VibeVoice-Realtime-0.5B`](https://huggingface.co/microsoft/VibeVoice-Realtime-0.5B)
- **Type:** Text-to-Speech / Voice Generation
- **Inference:** Real-time Streaming
- **Source:** Microsoft Research

---

## 🏗️ Technical Overview (From Original Repo)

<div align="left">

<h3>📰 News</h3>

<img src="https://img.shields.io/badge/Status-New-brightgreen?style=flat" alt="New" />
<img src="https://img.shields.io/badge/Feature-Realtime_TTS-blue?style=flat&logo=soundcharts" alt="Realtime TTS" />

<strong>2025-12-03: 📣 We open-sourced <a href="docs/vibevoice-realtime-0.5b.md"><strong>VibeVoice‑Realtime‑0.5B</strong></a>.</strong>

To mitigate deepfake risks and ensure low latency for the first speech chunk, voice prompts are provided in an embedded format. For users requiring voice customization, please reach out to our team.

</div>

### Overview

VibeVoice is a novel framework designed for generating **expressive**, **long-form**, **multi-speaker** conversational audio.
It addresses significant challenges in traditional Text-to-Speech (TTS) systems, particularly in scalability, speaker consistency, and natural turn-taking.

**[Realtime streaming TTS model](docs/vibevoice-realtime-0.5b.md)**: Produces initial audible speech in ~**300 ms** and supports **streaming text input** for single-speaker **real-time** speech generation.

<p align="left">
  <img src="Figures/MOS-preference.png" alt="MOS Preference Results" height="260px">
  <img src="Figures/VibeVoice.jpg" alt="VibeVoice Overview" height="250px" style="margin-right: 10px;">
</p>


### 🎵 Demo Examples

**English**
<div align="center">
https://github.com/user-attachments/assets/0967027c-141e-4909-bec8-091558b1b784
</div>

**Cross-Lingual**
<div align="center">
https://github.com/user-attachments/assets/838d8ad9-a201-4dde-bb45-8cd3f59ce722
</div>

For more examples, see the [Project Page](https://microsoft.github.io/VibeVoice).

## Risks and Limitations

While efforts have been made to optimize it through various techniques, it may still produce outputs that are unexpected, biased, or inaccurate.
**Potential for Deepfakes and Disinformation:** High-quality synthetic speech can be misused. Users must ensure transcripts are reliable and avoid using generated content in misleading ways.
**Non-Speech Audio:** The model focuses solely on speech synthesis and does not handle background noise, music, or other sound effects.

We do not recommend using VibeVoice in commercial or real-world applications without further testing and development. This model is intended for research and development purposes only. Please use responsibly.

## Star History

![Star History Chart](https://api.star-history.com/svg?repos=Microsoft/vibevoice&type=date&legend=top-left)