I've only remixed these four tracks with "AI"' as an exercise in what can be done with free software. I was curious. I'm not curious any more. It's not the fault of the app, I just don't see the point of "someone famous" singing a composition of mine. As far the @DEGRAD Miley Cyrus video as an example of what I lack remains somewhat obscure to me. If I don't write great melodies, or have great hooks, the most polished arrangements, the best mixes or mastering....I'm not worried. My friend writes the lyrics which reflect his struggles with a terminal illness. I have been friends with him since 1963 and he was a good guitarist, but due to his condition, cannot play any more, so I write his music now.

 

It's also a very new app, in the greater scheme of things. How would you rate your results using AI/clone options, as compared with the Dreamtonics synth you mentioned? Also the workflow aspects of it, etc. Synthesizer V is only 80$ now; for some reason I was under the impression it was much more expensive.

 

Hi, so can we install Replay on a W10 PC offline then download a few "voices" from https://www.weights.gg/fr ? Thank you.

 

[ Miley Cyrus vocal has a lot of "human imperfections" ] ..... hhahahaha , that cow aint nuttin but imperfection !!!

 

Last edited by a moderator: Nov 12, 2023

I also wanna know. I ask the same question in the main thread 2 days ago and still no answer

 

@zib @good! Download all of it and use a symlink so you can put it where you want , it worked for me.

 

Hmm.. so I cant install everything where I want?

When you say download everything, I already got the app and downloaded 5 models, what am I missing? Also, I am just confused on how it all works, how to actually create vocals and instrument sounds that actually work in my projects, do I type in words and/or use a mic to sing lyrics, melody and then replace/clone and render? As its standalone only, how to make the results in key with my song, synced to BPM etc. ?

As I said, I asked the developer in that other thread and got no reply.

 

I just read a comment on the main thread, it seems this app MUST be connected online to even complete installation and work. I dont know why the developer wouldnt provide a simple download link to those "initial default dataset models" so we can download/transfer for installing on an offline PC (or better yet, just include it with the app download, one and done!). OH well.

What is the file size of this "initial default dataset models" anyway?

I totally agree with the user comment below...

User Comment:

I think there's a major misunderstanding here. My DAW machine is offline, so I'm not willing to
let neither yours, nor any other program connect to the internet. Therefore in order to use Replay I need direct links to the 18 files it tries to download at startup, so that i can download them manually, as well the information where to put them on the offline machine, once downloaded.

In case you can't or don't want to provide these links, please consider removing the requirement of presence of obligatory default voice(s) from the program, and let users decide which voices they want use and let them choose the method of acquiring the voices.


Reply from developer:

No worries! The use case for our app isn't for you!

The initial models are actual models to run the app. The way the algo works is it needs the initial models to even run. They are the guidelines of how the human voice is comprised.

A lot of these AI tools need trained models to even run.

Then the user can put in the models they want that refer to the default dataset models that are the scaffolding for the python algo.

 

no I used it offline. everything you want I mean, of course the necessary files then at least one optional like Frank Sinatra or whatevere.

 

You use it offline, but you installed it online, is that right? Because according to the developer, it requires internet connection to install the 18 "factory models" for the app to even work - without these it wont work at all.

Where can I download the 18 mandatory "initial default dataset models" to transfer to my offline computer for installation?

Or do i have it all wrong and the developer is just being daft about providing information and download links for these essential files? All this back and forth, lack of answers or clarity and instructions is frustrating.

Apparently, there is a tutorial and VST in the works. YAY!

 

Last edited by a moderator: Nov 12, 2023

Yes you can

 

I am the developer. Please shoot any other questions to our Discord or my original forum post (https://audiosex.pro/threads/free-a...ning-app-with-built-in-stem-seperation.72525/)

we actually do have a VST/AAX/AU but having a lot of issues with it. And from what I have noticed from the community I want to make sure things are ready to give to everyone on here before people tell me its a "waste of time" app I am developing.

As for the initial download its not actual models to use in the app. It the AI algorithm models. It needs those to even run.

I should have put in the app "downloading essential files"

 

Last edited: Nov 12, 2023

This is exactly one of the many great use cases of this app!

But we can get even more creative. You can blend models with the app too.. so try to mix 60% elton jon with 40% of another anon voice and create your own sound.

 

Any tool in music-making is all about how you are using it. You can do amazing creative things if you expand your mind with how you use this!

 

Im sure this person said the same when a drum machine was invented. "this will be the end of human drummers"

fast forward to countless new genres that were created by incorporating the use of this new technology.

 

Yeah its a bigger vision other than using it for famous singers.

You can use the vocal models to do backup vocals.
You can blend multiple models together and create a new artist
You can use the instrument models as well


This is the early days of this app and the tech. Im sharing it with the community to come up some great use cases for the tech

 

I can't stand human drummers, they are evil people, I hoped until the end that they would be replaced by drum machines: unfortunately it didn't happen.


At the beginning of the industrial revolution (early 1800) a reactionary movement called “the Luddite” was created: they protested against manufacturers who used machines in a fraudulent and deceitful manner to replace the skilled labor of workers. That's pretty much what some people would like to do with AI now.


https://en.m.wikipedia.org/wiki/Luddite


My point of view is exactly reversed, I find AI still disappointing; it still fails to do what we expect it to do. I sincerely hope that they soon invent a scanner for my brain, in order to bring out all the music that plays inside me and that cannot come out with known methods.

 

I like that one. I'd check this thing out but cannot install on High Sierra. Oh well, it's not like I paid for it and have some right to complain.

Just like I wouldn't expect a software developer to acquiesce to my demands for an Offline Installer method, so it was easier for me to crack. But I would like the system requirement for MacOS to be 10.13.6, if it is not utilizing GPU.

 

I still use the Dreamtonics app primarily. But it's voice databases are not at all nuanced. If you select one singer, it will deliver one performance with the same inflections every time. Replay, at least allows some randomness into the result. I'm not sure how I feel about using it moving forward. Over the last two years I had created about 80 songs where I sang. Then 6 months ago I switched to Synthesizer V and have added another 30 songs. I have some misgivings about producing perhaps the next 30 with a roster of artists (living and dead) picked from the last 50 years of Grammy Award shows.
I don't suppose I should care really. I just do it for my own entertainment.

 

Your wish for a scanner is drawing closer to fruition thanks to these technological advancements. These tools are not the end of the road; they are merely stepping stones on the path to solving a larger puzzle.

Electricity had to be invented before the lightbulb could be created. Even after the lightbulb's invention, there was debate over whose patent was submitted first.