Why not open source 😐
So then why was the model given? If the fine tuning and voice cloning is all closed, then this model is useless. There are better open source models than this which can be used. 🙄
Hi @Deathgod7890 We hear you and we truly appreciate your feedback. There are many factors that influence what capabilities we can release publicly. We will keep the community updated as we progress.
back to rent-seeking 🙌
Hi @Deathgod7890 We hear you and we truly appreciate your feedback. There are many factors that influence what capabilities we can release publicly. We will keep the community updated as we progress.
This is lame. Not so much that it isn't an OSS license but that CC NC is the lamest of all. Why not add a license that at least allows developers to get a small user-base being served (Say, 1k) before they have to start paying? At least that way I'm not breaking the law by integrating your model in my product that has a handful of users. Nobody wants to get sued for some dumb shit. Nobody wants to use your API when this super small model weights exist for the same thing.
Hi @bitnom , The CC BY-NC 4 came from the voice in the dataset. We cannot remove it. Hope this help.
License
The provided voice-references compatible with this model are licensed under CC BY-NC 4, e.g. from EARS, CML-TTS, IndicVoices-R and Arabic Natural Audio datasets. Thus, this model inherits the same license.
There are many factors that influence what capabilities we can release publicly.
Translation: we like money 🤑
Hi @evewashere , One thing to keep in mind is license and compliance can vary significantly for AI labs in different countries.
Btw, We put every detail you need to know for training this model into the paper: https://arxiv.org/pdf/2603.25551. Hope this help!
If you think "fine tuning and voice cloning" is the only thing that is not available you are in for a laugh,
the whole release does not work at all haha, they purposefully choose the least compatible provider.
I actually have not heard anyone not affiliated that got it running, its all a smokescreen. do not waste your time on this release,
you will regret it.
yea this entire release is broken and useless
If you think "fine tuning and voice cloning" is the only thing that is not available you are in for a laugh,
the whole release does not work at all haha, they purposefully choose the least compatible provider.I actually have not heard anyone not affiliated that got it running, its all a smokescreen. do not waste your time on this release,
you will regret it.
yea this entire release is broken and useless
There only actual "issue" is that they can't use your work commercially for free, I'd wager.
There only actual "issue" is that they can't use your work commercially for free, I'd wager.
this you?
@y123456y78 , if I use a different voice by making my own voice encoder, does CC BY NC still apply to the model?
its a piece of shit,just use omnivoice