9 351
subscribers
About the creator
Hi all!
We have created a unique online application for separating songs into vocals and music using neural networks. Our MDX and UVR models are trained using our own dataset, which includes music of various genres.
High quality is ensured by resource-intensive algorithms, which we use powerful and expensive servers. Your financial support helps not only to pay for hardware rental, but also to develop the project further.
We offer both free and premium versions to our users. In the second case you get:
1. Bigger limits (depending on the subscription level). The more audio minutes, the more songs you can process.
2. WAV / FLAC downloading.
3. Backing vocals extractor.
4. Piano and strings models.
As an experiment, these premium benefits will be available to all users for a while:
1. Better quality processing with the latest models.
2. Separation into lead vocal, bass, drums, guitar or saxophone.
If you are a new subscriber, you will receive a link in a private message immediately after payment. Or you can find it here:
Standard → https://boosty.to/uvr/posts/f8228831-dfb4-4157-b644-5e4a1d6dbd59
Professional → https://boosty.to/uvr/posts/29188ef6-e5ab-4878-a287-34fcb5279dde
Ultimate → https://boosty.to/uvr/posts/4aea31e3-87c5-42e6-86bc-5244c6153b27
Professional → https://boosty.to/uvr/posts/29188ef6-e5ab-4878-a287-34fcb5279dde
Ultimate → https://boosty.to/uvr/posts/4aea31e3-87c5-42e6-86bc-5244c6153b27
If the payment has failed, try these tips:
New Mel-RoFormer vocal removal model
Introducing the new version of Mel-RoFormer, which can really be called the best vocal removal model in the world. This is a huge step forward from the previous version! The new model removes vocals less aggressively, reminiscent of the sound of the old UVR models, but it does it in a much better way.
And for premium users, "correct_phase / de-noise" post-processing is available to make the music sound more natural.
Show more
Show more comments
Ktmrocky
Old Mel-Roformer (kim) is best for me ❤️
Nov 05 14:29
1
B3'N
Both seem pretty good. Thank you ^_^
Nov 08 01:41
Similarity / Phantom Center Extractor model
Added a center channel extraction model. It is very similar to the Center Extract effect in iZotope RX, but works better.
It separates the song into two tracks:
mid refers to the phantom center, meaning the part that contains instruments and vocals perceived in the center.
side refers to everything perceived in stereo.
You can find this model in the Extract Backing Vocals section. Although it does not isolate vocals, it can be useful in your work.
Example song:
Loveshadow - I.E.
UVR Online
Loveshadow.mp3
0:00
0:25
Mid:
UVR Online
Loveshadow (Mid) (Phantom Center Extractor).mp3
0:00
0:25
Side:
UVR Online
Loveshadow (Side) (Phantom Center Extractor).mp3
0:00
0:25
Weekly report
1. In the new interface, the BS-RoFormer model now also has De-mudder.
2. We have started work on two new models: RoFormer De-reverb and De-echo. We are also experimenting with a new model for backing vocals isolation.
Show more comments
cristian gonzalo guerrero
es imposible escuchar los resultados en la escucha previa del reproductor de la web se traba siempre y no se puede hacer nada!!!!!!!!!!!
UVR Online
cristian gonzalo guerrero, Póngase en contacto con nosotros a través de mensajes privados.
Creator has disabled comments for this post.
Subscription levels
Lite 2024
$ 2,9 per month
330 min of audio per week
Max song length: 15 min
Standard
$ 4,8 per month
750 minutes of audio per week
Max song length: 20 min
heavy models - 15 min
Professional
$ 9,9 per month
Great for frequent use
1900 min of audio per week
Max song length: 25 min
heavy models - 15 min
Ultimate
$ 21,1 per month
600 min of audio per day (past 24h)
Max song length: 30 min
heavy models - 15 min