1-Click Vibevoice-Realtime for ALL machines Finally a Realtime TTS that is ACTUALLY realtime, thanks to its tiny size! (0.5B params) This video is from my Windows machine, recorded in realtime. All you need is around 2.5GB VRAM, it even works on Macs!

Run it on your local PC with 1-click, available via

Fun fact --- This 1-click launcher was 100% built by AI in Pinokio!

Official Github from Microsoft:

The launcher automatically: 1. Downloads the new VibeVoice-Realtime-0.5B model from Huggingface 2. Sets up environment. 3. And launches the app, all with one click.

Just a quick note: It's not perfect yet. You may notice a slight choppiness as it starts speaking. I would like to think that it's not the model's problem but just this specific code implementation (The official github code), hopefully this becomes smoother.

Sounds robotic

Set this up today!

So wild ! Love it. Just played with it. Had it up and running in couple Clicks

The real fun begins when you start remixing the app using AI. Do play around!

Will it work on mobile devices as well?

Need to look into this. Does this model support voice cloning or a fixed set of voices?

i think it's a fixed set of voices, i could be wrong but i think the voices can be trained. let me know if otherwise!

Nice work. Real-time TTS with a tiny model. Curious how it handles longer texts and streaming latency. The AI-built one-click launcher angle is cool.

sounds like sh!t

@grok is this 100% offline?