Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

VibeVoice: Open-source frontier voice AI

github.com

Post reply on HN
Print view View original

181–190 of 191 posts

Re: VibeVoice: Open-source frontier voice AI

#181

Post by Vinnl »

fg137 wrote: post #116
Vinnl wrote: post #69

Earlier quoted context omitted.

"get offended" is just what the clickbait news cycle made of it. It was based on the post at [1], and this is all it said: > We need to get beyond the arguments of slop vs sophistication and develop a new equilibrium in terms of our “theory of the mind” that accounts for humans being equipped with these new cognitive amplifier tools as we relate to each other [1] https://snscratchpad.com/posts/looking-ahead-2026/

Are you sure you have the correct reference? I think everyone else is relating to https://futurism.com/artificial-intelligence/microsoft-bans-...

Heh well, that article says it "clearly infuriated executives at the company", and links to [1], which is exactly what I described. But banning it on Discord does kind of retroactively prove their point, I suppose.

[1] https://futurism.com/artificial-intelligence/microsoft-satya...

Re: VibeVoice: Open-source frontier voice AI

#182

Post by d4rkp4ttern »

This was on HN 7 months ago:

https://news.ycombinator.com/item?id=45114245

Every time a STT/TTS model is posted I wonder if it will change my current workflow on MacOS, which is:

STT with Parakeet-V3 via Hex [1] app for near-instant good-enough transcription for talking to AI agents.

TTS using KyutAI’s Pocket-TTS, an amazing 100M-param model. I used this to make a "voice" plugin [2] for Claude Code

So far I haven’t seen anything that replaces these for me, or haven't been persuaded enough to spend time testing an alternative (explore/exploit and all that).

[1] Hex STT app - https://github.com/kitlangton/Hex, which is macOS-only. (also good free/OSS alternatives: Handy, VoiceInk. No need for Wispr, Superwhisper etc)

[2] Claude Code Voice Plugin - https://pchalasani.github.io/claude-code-tools/plugins-detai...

Re: VibeVoice: Open-source frontier voice AI

#184

Post by simonw »

LtWorf wrote: post #180
simonw wrote: post #100

Earlier quoted context omitted.

The OSI's take on this is that an open source model can be modified through fine-tuning etc, even if you can't rebuild it from scratch. The problem with requiring "build from scratch" for open source models is that the number of interesting models with training data that can be openly licensed is close to zero. If you trained your model on an unlicensed scrape of the web you can't release the data under an open sourc…

I think the OSI no longer has any authority since that stunt they pulled in their "elections".

I'd missed that whole thing. Useful context: https://lwn.net/Articles/1014603/

Re: VibeVoice: Open-source frontier voice AI

#186

Post by MarsIronPI »

LtWorf wrote: post #179
MarsIronPI wrote: post #168

Earlier quoted context omitted.

I see your point and I agree. It's just that when you say "GPLv3 says that you can only use the software if it's either a cloud service, hypothetical open firmware devices" that's a stretch and not really true. AIUI vendors can pre-install GPLv3 software as long as they let you actually then replace the software (i.e. no DRM or locked bootloader). The firmware can still be non-GPL and non-replaceable. You just can't…

AFAIK you can use GPLv3 for non-replaceable stuff. The thing is only to allow the users to replace it IIF it's phisically possible to do so. If you make a device that boots from a ROM it's not a problem. If you sign your updates and keep your public key on a ROM and there is no way to boot anything else… there's a problem.

> If you sign your updates and keep your public key on a ROM and there is no way to boot anything else… there's a problem.

As there should be.

Re: VibeVoice: Open-source frontier voice AI

#187

Post by steinvakt2 »

d4rkp4ttern wrote: post #182

This was on HN 7 months ago: https://news.ycombinator.com/item?id=45114245 Every time a STT/TTS model is posted I wonder if it will change my current workflow on MacOS, which is: STT with Parakeet-V3 via Hex [1] app for near-instant good-enough transcription for talking to AI agents. TTS using KyutAI’s Pocket-TTS, an amazing 100M-param model. I used this to make a "voice" plugin [2] for Claude Code So far I haven’t s…

What do you consider to be the model with highest accuracy?

Re: VibeVoice: Open-source frontier voice AI

#188

Post by steinvakt2 »

isodev wrote: post #125

I think in this category, Voxtral by Mistral is a lot better. It also happens to be small enough to run on webGPU https://huggingface.co/spaces/mistralai/Voxtral-Realtime-Web...

Better accuracy than whisper large? For English? What about multilingual?

Re: VibeVoice: Open-source frontier voice AI

#189

Post by d4rkp4ttern »

steinvakt2 wrote: post #187
d4rkp4ttern wrote: post #182

This was on HN 7 months ago: https://news.ycombinator.com/item?id=45114245 Every time a STT/TTS model is posted I wonder if it will change my current workflow on MacOS, which is: STT with Parakeet-V3 via Hex [1] app for near-instant good-enough transcription for talking to AI agents. TTS using KyutAI’s Pocket-TTS, an amazing 100M-param model. I used this to make a "voice" plugin [2] for Claude Code So far I haven’t s…

What do you consider to be the model with highest accuracy?

I guess you mean for STT. For my usecase of talking to AI's or coding agents, pure STT accuracy is less important than transcription speed. Transcription needs to be near-instant, and accuracy "good enough" so that the AI's can "read between the lines". Parakeet-V3 gives exactly this.

Re: VibeVoice: Open-source frontier voice AI

#190

Post by SyneRyder »

gagan2020 wrote: post #93
steinvakt2 wrote: post #10

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.

It is not good for text to speech (TTS) as well. I am trying it for few days. First of all 1.5B model documentation is not there. 0.5B realtime is shit model. I was converting text, line by line and it was randomly adding music and couldn't handle special characters like "…". I really disappointed with this model to say the least.

> ...it was randomly adding music...

I've been noticing this with the Mistral Voxtral TTS models too. I have my AI record a morning briefing podcast for myself, and occasionally there are sounds like music at the start (the british voice had a musical tone underneath that sounded a little like the end of the BBC News theme). I don't think I've ever encountered that with the OpenAI TTS models, so they're now my default go-to again.

Post reply on HN