VibeVoice: Open-source frontier voice AI

#1

Post by tosh » Tue, Apr 28, 2026, 11:56 AM UTC

VibeVoice: Open-source frontier voice AI
github.com

Re: VibeVoice: Open-source frontier voice AI

#2

Post by CubsFan1060 » Tue, Apr 28, 2026, 12:44 PM UTC

Great post last night from Simon: https://simonwillison.net/2026/Apr/27/vibevoice/

Re: VibeVoice: Open-source frontier voice AI

#3

Post by podgietaru » Tue, Apr 28, 2026, 12:45 PM UTC

So we've really just settled on Vibe as the verb for AI then?

Re: VibeVoice: Open-source frontier voice AI

#4

Post by pryanshu89 » Tue, Apr 28, 2026, 12:51 PM UTC

podgietaru wrote: post #3

So we've really just settled on Vibe as the verb for AI then?

Why use precise technical language when you can just vibe with your AI system?

Re: VibeVoice: Open-source frontier voice AI

#5

Post by embedding-shape » Tue, Apr 28, 2026, 12:52 PM UTC

Isn't this project the one Microsoft published but then soon after pulled it for security/safety reasons? What has changed since then?

Re: VibeVoice: Open-source frontier voice AI

#6

Post by giarc » Tue, Apr 28, 2026, 12:53 PM UTC

podgietaru wrote: post #3

So we've really just settled on Vibe as the verb for AI then?

I'd be willing to bet it will be "Word of the Year" for 2026. Merriam-Webster had 'slop' for 2025, and 'polarization' for 2024. Is there a prediction market for this?

Re: VibeVoice: Open-source frontier voice AI

#7

Post by 542458 » Tue, Apr 28, 2026, 12:53 PM UTC

CubsFan1060 wrote: post #2

Great post last night from Simon: https://simonwillison.net/2026/Apr/27/vibevoice/

Note that this just covers the Speech-to-Text/Speech-Recognition aspect (a-la whisper), there's also models for long-form Text-To-Speech and steaming Text-To-Speech.

Re: VibeVoice: Open-source frontier voice AI

#8

Post by walthamstow » Tue, Apr 28, 2026, 12:54 PM UTC

Seems quite heavy for a STT model, Parakeet and Whisper are much smaller and perform great for quick dictation and transcription of longer files. I guess that's due to additional accuracy and speaker diarisation?

The TTS example clip in the repo of 'spontaneous singing' is creepy as fuck

Re: VibeVoice: Open-source frontier voice AI

#9

Post by 542458 » Tue, Apr 28, 2026, 12:55 PM UTC

embedding-shape wrote: post #5

Isn't this project the one Microsoft published but then soon after pulled it for security/safety reasons? What has changed since then?

Look at the "News" section in the readme - The original TTS model is gone from this repo (you can still find it other places), but the SST/ASR, long form TTS, and streaming TTS models are newer.

Re: VibeVoice: Open-source frontier voice AI

#10

Post by steinvakt2 » Tue, Apr 28, 2026, 1:00 PM UTC

This is not a new model. Also, it hallucinates a lot. Also, it's very heavy and slow in inference. It's also bad in multilingual.

Edit: I'm talking purely about speech to text (STT). Not sure about the other things this can do.