Відео

Hello, Thank you for the comment! Please feel free to try out the trial of VoiceAI that is available in your Sonarworks account. After evaluating the software you can see if it suits your professional needs.

I don't mind to pay someone who bridged RVC into my DAW in form of plugin and convenient workflow. And trained set of great models. What's wrong with you people, you just have no life if someone trying to make money turn you that mad

So what? I'm happy someone made a convenient plugin out of RVC finally. Hope they will allow training/loading custom models soon

I want to dissapoint you, mostly every feature you use in DAW or plugin is based on the open-source libraries/tech. Go ahead, make your own plugin and give it away for free. Maybe you will reach the bright future of communism one day

I don't mind to pay someone who bridged RVC into my DAW in form of plugin and convenient workflow. And trained set of great models. I'm happy someone did this finally

@@DorbyMorrow-lj6fg Hi Dorby! Let me remind you please, that RVC is monophonic ONLY, meanwhile this video shows the models supposedly outputting Stereo, which is impossible using HiFiGan, which RVC is built off of. Secondly, these voice models are actually very bad, and sloppily trained. Finally, there's so much messing around with the audio and so much "editing magic" that I doubt he product even works as they claim it does, as they show the models loading and converting instantly, which is impossible, as the models need time to initialize and load into memory, which takes around 10 seconds. Hope this helps clear some things up about this product which attempts to paywall RVC, something many other companies have tried, and subsequently failed, to do!

I don't agree with you about models quality. I tried A LOT of RVC models, trained my own.. and have to admit Sonarworks did a good job training these ones. Just a time saver. That would be nice if they add opportunities to train our own or blend existing ones though

@@toksik_protogen Why are you REALLY so angry about this plugin?

@@DorbyMorrow-lj6fg I am a verified Model Maker for RVC who makes high-quality models, these models sound absolutely atrocious compared to a genuinely good and high-quality model

Hello, Thank you for reaching out! Yes, the VoiceAI plugin will be working with any other language as it will only alter the sound of the voice itself. I hope this helps!

Hello, Thank you for reaching out! The frequency curve can always be adjusted if required. I hope this helps!

Is it a subscription? It says on the website that you pay for credits. That's not the same thing as a subscription. The trial version is just free credits. Did I miss something?

Hello, Thank you for reaching out! The processing time it takes for the SoundID VoiceAI to process your captured audio into the target voice and instrument is paid by tokens. Here's an example of tokens spent in a specific scenario (vocal replacement for a full song): 1. Capturing a 12 seconds sample audio of a voice track 2. Processing the sample with 5 voice presets and trying 3 different Transpose settings on each preset to find the best fit: 12x5x3 = 180 seconds / 3 minutes = 1800 tokens 3. Processing the entire vocal track of 2.5 minutes = 1500 tokens 4. Total processing time and token cost: 5.5 minutes = 3300 tokens For more details on the token system, see our guide here: soundid-produce.support.sonarworks.com/hc/en-us/articles/17399357472402-Tokens-and-minutes Let us know if you have any further questions!

Hello, Thank you for the comment! Unfortunately, SoundID VoiceAI is a plugin only and is not available in standalone mode. I hope this helps!

so true, and they have now many voice to work with. and support ARA technology to work directly in DAWs

Edit the performance in real time? You can sing it however you like and sound like someone or something else. Some people are in denial and it shows

They also have audio to midi or visa versa.

@@Edbrad you can also sing make midi with ARA and make other AI sing, you can edit vibrato, pitch, length, words, phonems etc...

Seriously? How is it boring? You’re just sore about AI. This is the future.

You should be happy this is a tool for producers. What’s the difference between this and using samples? People have been using samples for decades now. This is hardly any different

@@Edbrad ..And you should reflect more before asking such questions. So I should be happy, because I'm a producer? Maybe the term 'producer' should BE defined. Someone who sticks samples together in a daw is not a producer, even if Generation Z thinks so. So I should be happy when human resources are replaced by machines that are trained unpaid by artists' works in order to then replace them? Only someone who has no resources or skills of their own could be happy about this. Being a singer is also a job. These people have spent a long time learning to be able to do what they can do and how they can do it. What is the difference between Samples and AI? A person, an artist, was paid for samples, he was stolen from for AI and the value creation from his skills takes place without him receiving any benefit. Yes, if you don't think a millimeter further than for your own immediate benefit, you might be happy about it. But then you're also an ignoramus and that's nothing you could or should be happy about.

Not true. A really good singer will always be in demand. And for now you still need to be a good singer for these tools to sound their best. A bad performance that's out of tune and out of time can not be saved by voice AI.

Which one is free ?

@@ChamilawarnaJayalath replay!

Exactly This.

Hello, Thank you for reaching out! If there are any issues with SoundID Reference, please create a support request in our support center and We'll be happy to get things sorted.

Try it out, there's a free trial. I will warn you though, if you put in a terrible performance don't expect magic. But if you put in a decent performance this tool will polish it and make it stand out.

because right now producers have to leave the DAW and go to another platform to do this. Now it's a plugin. Why is that hard to understand 🤣🤣 60% of musicians have already used AI for their productions and its only going higher.

@@BenoniStudio I'd like to see where you get your sources and numbers from

Until next year, i bet

no, AI has already plateau'd. We're on top of the S curve. The hype won't last.

If you say 'not new, It exists for 2 years already', you would be on point. But 10 years... not really. GAN neural networks concept was presented in 2014, but convincing results with audio started appearing in 2018-2021. But in early prototyping form, as demo for research papers. Technology became available for everyone since end of 2022.. And there just few products in the market putting it into form of DAW plugin. I don't see SW is pretending to be inventors of this tech, they just claim to be early adopters of putting this technology into convenient and usable tool fo every creator

this is new because its a PLUGIN. try to comprehend things and not just shake your fist at the sky because you cant cope with a changing world. 🤣

​@@BenoniStudio LOL. I can cope with a changing world, and have worked with these effects for many years. Its not AI. Its just the Emperor Without Clothes again.

​@@so_you_can_go But you dont need AI to do these effects. So 10 years yes. It also exists as hardware and has existed for many years - and those manufactors dont call it AI either. So the two years you talk about, is probably the last two years hype of AI - a technology evolving fast now - but it existed in the 80s too. The first Hebbian network was simulated in 1954. Btw. I could do pitch shifting in the eigthies too and make ugly effects. If Sonarworks/SoundID used AI to generate these effects, I would expect it to be much better quality than these general effects. Some of the demos actually sounds like my old 80s pitchshifter. The others sounds like something I've been able to do for 10 years.

@@JX3D I tried the trial and would say this way: there is no pitch shifter being able to convert my voice into trombone, trumpet, bass guitar or conga. You'd be better to try it first before hating, otherwise you make yourself look bad

It sounds like you have a negative bias towards 'AI,' and you think that straightforwardly criticizing the tool, even if you don't truly believe in what you're saying, is going to make a difference somehow.

😂😅🤣😅😂😅🤣😅😂

They said the same thing about MIDI, and Pro Tools, and Auto-Tune, etc...

@@BenoniStudio I've never heard anyone say that about MIDI. But they did say it about Auto Tune, and many still do. And Pro Tools is not the nicest software to use, but it was one of the best tools when it was at its high - now all competiors have better software.

@@JX3D Pro Tools is used as an example of a DAW since it was essentially the first to be adopted, though it was called Sound Tools then. And if you look at the history of composers crapping all over MIDI you may understand this "AI SUCKS AND ITS NOT REEEAALLLL MUSIC MAAANNN" is what all people do when they are threatened by new tech.

I almost tried your sw for room correction, but it doesn't support umik1 because that's usb. So I didn't.

What is this discrimination against usb, btw ? XLR mics have to connect to a PC through an ADC the same way, don't they ? I suppose the audio interfaces are connected to the PCs by usb, don't they ? I don't understand it all.

@@gmlgml780 USB mics have different clock so input and output are always out of sync (phase drift).

@@reinislusis3569 _"... USB mics have different clock so input and output are always out of sync (phase drift) ..."_ Yeah. Audio interfaces connects through USB, too. Why do they not have "different clocks" ?

@@gmlgml780 because audio interface has same clock driving ADC and DAC. That is why more expensive audio interfaces have separate input/output for clock source. You might have some luck using USB microphone on macOS using aggregate audio device (but likely phase drift will cause some issues) however on Windows Sonarworks Reference Measure uses ASIO and requires input and output to share same ASIO driver. Regarding phase drift - each audio interface has internal clock crystal (44100 / 48000 / etc) however it is slightly imprecise (say 0.01% error). However even small error like 0.01% @ 44100 means 4 sample error per second which in minute accumulates to 240 samples or 1.85 meters or locating error.

Hello! Thanks for leaving a comment. Currently, RME devices are not supported by our platform. We encourage you to keep an eye on our updates. As soon as RME is included in our supported devices, we'll make sure to announce it to all our users.