it is an arm app. It used the MPS gpu for rendering.

the intel version is really buggy and slow.

 

you are the goat for this!! Thank you for the info! A fix to follow!

 

Rosetta is for non-ARM

 

Any idea how to fix this error off the top of your head brother?

 

Last edited: Nov 14, 2023

we just pushed an update with this feedback.. let me know if it works!

 

that’s a weird one. We just pushed an update try to reinstall that version and let me know if it fixes it

 

@curtified is there a way to reset all settings to initial factory settings. I've been toying with them to improve functionality and my results are just getting worse.

 

Update worked now. Thanks

 

Happy to see that! Thanks for your patients with us!

 

oh no!! Nothing for that right now but I will add it to a future build.

let me see what file to delete to be safe

 

I just tried installing this junk on a work Mac m1 with SOMONA, nothing works but playback of any audio or video the user loads from their desktop, I can download a voice but cant hear it or create anything with either file I add or voice I download..malware?

 

Last edited: Nov 15, 2023

Oh so you are still around here, I remember you posted a topic about a similar subject (AI generated voices) but then you posted about your departure from the forum.

 

Moving forward im not going to humor this type of feedback.

There has been respectful honest feedback in here. Problems we have solved together.

 

You nailed it, that works without renaming the intel dll file now, good job!
Thanks for the fix

By the way, any suggestion to speed up the stem separation process? A faster stem method among all?

 

Last edited: Nov 15, 2023

That stemming method seems to be the best sounding without doing ensemble mode. (which we will include in future updates).

I have to do timed tests of each algo to see which one is fastest over best quality for users.

The other option that I still use is pre stem out your audio using UVR (https://audiosex.pro/threads/uvr5-the-best-ai-stem-separation-algo.69463/) then import the dry vocal into Replay for vocal conversion.

 

nothing works on m1 somona Mac set up, nothing solved, server errors, download permission errors, even their discord see does not load in discord, I shared screenshots before and you claim things where solved, well, when and where? lol

 

learn a bit more about your Work Mac m1 with SOMONA ole mate instead of acting like a child with a silver spoon mentality. become a knowledgeable person instead of a cooker with the brain of a fart.
if you actually asked politely there are members here that would of gladly helped you but now after your outburst go google the information for yourself and learn to manually set the permissions if needed to allow the process's to work

and btw it does tell you in your image if you cared to look *sigh*

so as for your comment on this is junk, unfortunately all you did was display your intellectual incomptence good luck in your future.

 

Last edited: Nov 16, 2023

Great!

 

I am wondering in the future will you add a possibility to convert written text (text to speech) to a specific voice model, instead of adding audio file.

 

yes!

I am working with suno http://app.suno.ai to do just that!! Try it out!

This is just the early days. But let your imagination run wild.