tutorial icon indicating copy to clipboard operation
tutorial copied to clipboard

Tutorial covering Open Source tools for Source Separation.

Results 20 tutorial issues
Sort by recently updated
recently updated
newest added

Some kind of typo/rendering issue. Ctrl+F for: added \text{artif}acts Seen on OSX Safari 15.2.

"For example, we might want to isolate a singer from the background music to make a karaoke version of a song or isolate the bass guitar **from the from** the...

Fixed issue [#28](https://github.com/source-separation/tutorial/issues/28) in Training > Gradient Descent > Signs of Healthy Training Changed: `ax[1].plot(np.log10(losses[j]), label=f'Learning rate: {lr}')` to be: `ax[1].plot(np.log10([t.detach().numpy() for t in losses[j]), label=f'Learning rate: {lr}')`

We need to detach a tensor in the ["Signs of Healthy Training"](https://source-separation.github.io/tutorial/training/gradient_descent.html#signs-of-healthy-training) section on the Gradient Descent page. Here's the cell: fig, ax = plt.subplots(1, 2, figsize=(15, 5)) for j,...

bug

**What Section?** repetition.ipynb **Describe the bug** In the tutorial python, something gets messed up with the Gradio connection. When you call `repet.interact(share=True, source='microphone')` The following error gets thrown. > ---------------------------------------------------------------------------...

bug

I try to play with the colab notebook you provided. However, when I try to run ```python %%capture !pip install scaper !pip install nussl !pip install git+https://github.com/source-separation/tutorial %%capture from common...

In section: Exercise making it interactive, the `interact` function is broken, here is part of the traceback: > [/usr/local/lib/python3.8/dist-packages/nussl/separation/base/separation_base.py](https://localhost:8080/#) in interact(self, add_residual, source, label, ext, separate_fn, outputs, inline, inbrowser, share,...

In the code block for the MaskInference class, `connections.append([key, f'model:{key}'])` is missing brackets around `f'model:{key}`, which causes the code to run into a KeyError. The correct code should be `connections.append([key,...

Your issue content here. Seems like the link for tasnet and demucs have been somewhat mixed up.

Refs to Architectures sections are mixed up