Have you tried running the output out of Dreamtonics through some of the voice models in Replay? I would love to hear the results. Maybe it can give some character to how robotic DT can sound at times??

If so plz share here!

 

NO, you cannot. I finally deciphered the puzzle of installing this lol..

1. Download and install the app.

2. Upon opening for the first time, you MUST download the "factory" models content (18 files 5-8 GB) which requires internet connection to do so. The app will not work without this content installed.

3. Download and install the "user" models you want to use.

4. NOW you can turn off that pesky internet connection and use the app offline (on the same machine).

*If your PC is totally offline, you wont be able to complete installation. You MUST be connected to the internet to install the mandatory 18 factory model files required for operation.


@curtified, youre welcome, in less than 1 minute I wrote your installation tutorial for you ;)

 

Last edited by a moderator: Nov 12, 2023

Thank you for condensing all of my answers throughout this forum into one place, and for taking the time to try the app yourself. I hope that by trying it out, you were able to answer some of the questions you had in your initial post on the other thread!

Now I can focus on fixing the grocery list of bugs in this code I've been working on!

 

@ curtified ..... yer a funny person , and its very odd you make your statementsand , really an aasumption about me in here
when you dont know who I am. Im thinking you must have absolutely no talent when it comes to music or even playing
a musical instrument , am I right ??

 

Where to download the "factory" models content of best singers to start using replay....

Any repository or cloud package link...
Even a website with easy download the "factory" models content of best singers (male and female)

 

Okay..... this is the wav file output from a singer in Dreamtonics Synth V directly into Replay using the Miley Cyrus vocal model. I don't hate it, so keep up the good work!

 

I understand what you say and I completely agree: technological progress is a constant flow, just think of the evolution of cell phones from the beginning of the 90s to now. What I find boring about AI is all this boasting for no reason: at the moment it doesn't seem like such a useful tool. We have all had the opportunity to test apps dedicated to music production:

• ChatGPT to write lyrics
• Neural net to create chord progressions
• Voice cloning
• AI plugins
• AI mastering

(For those who still don't know the topic in depth, try google for "AI for music productions" and try it yourself)
Unfortunately, there is no magic in these apps/webapps/plugins: the results are rather disappointing, I haven't found anything that amazes me yet. I sincerely hope that this technology will give us (quickly) more concrete results

 

This is dope!! First fully AI + voice clone example I’ve ever heard!!

 

All the models are at www.weights.gg

 

It’s actually the exact opposite.

This is my Music page
http://www.SoundCloud.com/Flosstradamus

 

All of these things can be co-pilot tools.I think people see AI and think they will do the entire job for them.

Instead of expecting these tools to do the entire process find ways to work with them. Here are some random examples.
- work with ChatGPT to help you find 10 options of words to rhyme with what you're stuck on.
- have NN help you complete a Chord progression in a new way than you would usually go
- use multiple Voice cloning to sing a choir of different backup vocals so support your lead vocal
- Use gullfoss to pre mix your channel before applying other EQ etc
- have newfangled elevate do some surgical work before you apply your main limiter

 

This way I work twice, doubling production times. Currently, it seems that AI, instead of solving my problems, creates new ones. AI for music production is, as I said, disappointing.
An example: I'd like to play a track for Frank Sinatra, where I provide lirycs and midi files and the app brings out Sinatra's real voice following my notes. AI marketing claims that apps can do this, but in reality it is not true. No prob, I have faith in technology and engineers: I'm sure we'll get there soon.
@curtified , we are in your hands

 

we will be there soon for sure!!

I've been advising on this project that does text prompt to full song generation.

http://app.suno.ai

Its still under a minute in render length and 128kbps. This is def a step in the right direction of what you are looking for. Try it out! Think about this tech in 1-3 years!

 

I told it to to make a “jazz swing style song about a online forum called audio sex” lol

https://cdn1.suno.ai/2f019d00-90c2-4b51-8bb3-6d14789fb087_1.mp4

and

https://cdn1.suno.ai/2f019d00-90c2-4b51-8bb3-6d14789fb087_0.mp4


Now I know on this day in 2023 it’s not there. But from a few words I typed in that’s a pretty good attempt at a fully generated song!

 

Ok so I spent 30 mins & used Suno + Replay + Gullfoss + TEOTE to make a crappy example of a full AI generated song by Frank Sinatra where it talks about Audio Sex

https://voca.ro/1gj6tUKdIlmt

Remember this is the worst this tech will ever be.

 

This model sounds much more accurate than the others. Maybe it's not perfect, but you can tell exactly who it is supposed to be. It's about as close as an Elvis impersonator.

 

I agree it’s not 100% (yet)

and I will respect any feedback from someone with the name clone when it comes to cloned voices:

 

Hmmmmmm , well I guess its 'music' , but mostly all I hear is machine droning loops and push button - knob twisting gadgets. Please , tell me , you dont think this is what make you a musician , because your really NOT !!

 

For another AI tool that is absolutely killing other modern plugins you should all test out Hush.AI, it's a noise reverb remover for dialogue and in testing it destroys my usual plugin chain and the type of AI tool we need. I have zero to do with Hush.AI, I just thought I would share my opinion as in testing it has really upped the bar for removing noise and room sound from vocals in my usage.

 

Last edited: Nov 13, 2023

Im not a musician. I’m a hacker. I legit make music like a hacker.

I will take elements from different sources. Samples, midi packs, presets etc. find ways to hack those parts together to make the creations I put out on music platforms.

thats why we built Replay. I needed a tool to hack together human voices and instruments in a new creative way.

i approach music, tech, art, the world all with the same vantage point.

As for you not likening the music I make I’m ok with that. I made all of it for me. I like it! I really enjoyed making it so I thought I would share my experiments with the world! And it appears that others like it too!