francesclluis / source-separation-wavenet Goto Github PK
View Code? Open in Web Editor NEWA neural network for end-to-end music source separation
License: MIT License
A neural network for end-to-end music source separation
License: MIT License
I did exactly what the guide said but nothing happens
Hi.
I was wondering how I can use GPU for training?
I checked with tutorial theano python code and confirmed it uses GPU.
However, when I run wavenet, it does not use GPU and instead uses CPU.
Could you share how I can use GPU?
Thanks
The code uses python2 but the environment created from the file uses python3
Have you tried with up-to-date tensorflow? Does it support large dilation?
Hey!
First of all I want to thank you for publishing this awesome work.
In the readme you mention, that this is capable of doing real time source separation when run on a GPU. I'm really really interested in that use case.
However I can only find ways to give it input wav files and no way to tell it to use i.e. an audio device from the PC.
I'm assuming that this functionality hasn't been implemented yet. Would it be trivial to do so?
Judging by https://github.com/francesclluis/source-separation-wavenet/blob/master/separate.py#L82 there is some kind of minimal amount of samples one would have to give the network.
Thus would it be feasible to say pass it every new frame of sound data with the last few frames also attached to make it work on an audio stream?
Or is there a way to maybe feed to the network frame by frame? that would be awesome because then one wouldn't have to deal with stitching the different results back together which would probably result in some quirkiness.
I would be happy to contribute this feature, but want to make sure it's possible first :D
The correct line to retrieve the data from the dataset after extraction is
mus = musdb.DB(root=self.path, is_wav=True)
instead of
mus = musdb.DB(root_dir=self.path, is_wav=True)
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.