Interesting story about this repo/product/author by cybersecurity researcher Kevin Beaumont: https://cyberplace.social/@GossiTheDog/116454846703138243
VibeVoice: Open-source frontier voice AI
141–150 of 191 posts
Re: VibeVoice: Open-source frontier voice AI
#142This 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.
I'm shocked, shocked to find that Microsoft takes credit for a slow, unoriginal product that doesn't actually do what it advertises.
Re: VibeVoice: Open-source frontier voice AI
#143Shouldn't it be called something like "Copilot Voice"?
Re: VibeVoice: Open-source frontier voice AI
#144Earlier quoted context omitted.
> I care that I know what I can DO with the project when I see it described as "open source". Yes, the first of which is that you should be able to build it from source. Which requires the source code, and in this case data.
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…
By this definition almost any binary can be "open source" since hex editors exist. (Or more usefully, you can use ghidra et al. to do more interesting changes.) I know GPL has a very specific view of things, but I'd like to quote an excerpt that I think is generally applicable from https://www.gnu.org/licenses/gpl-3.0.html -
> The “source code” for a work means the preferred form of the work for making modifications to it. “Object code” means any non-source form of a work.
Which is why I'm fine with "open weights", because that's saying the object code is under an open license.
> 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.
So? If the number of open source models is zero, then the number of open source models is zero.
Re: VibeVoice: Open-source frontier voice AI
#145This 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.
Re: VibeVoice: Open-source frontier voice AI
#146Re: VibeVoice: Open-source frontier voice AI
#147Earlier quoted context omitted.
you saved us a lot of time here.... i unstarred the repo moving on....
I don't really pay attention to stars. Do people use them as bookmarks? Why would you star a repo if you knew so little about it?
Re: VibeVoice: Open-source frontier voice AI
#148Re: VibeVoice: Open-source frontier voice AI
#149Earlier quoted context omitted.
you saved us a lot of time here.... i unstarred the repo moving on....
I don't really pay attention to stars. Do people use them as bookmarks? Why would you star a repo if you knew so little about it?
Re: VibeVoice: Open-source frontier voice AI
#150Earlier quoted context omitted.
you saved us a lot of time here.... i unstarred the repo moving on....
I don't really pay attention to stars. Do people use them as bookmarks? Why would you star a repo if you knew so little about it?