Juan Coria

You can measure the parameter size of the diart pipeline by adding the sizes of the different models involved. For the segmentation and embedding models you can use any method...

Hi @KentDesreumaux, What devices are available in your raspberry? I'm guessing the "default" device may be the wrong one. Also, could you verify what's the input and output of `SegmentationModel`?

**Update:** pyannote.audio and pytorch installations already automated in v0.6 I prefer to leave a pyannote installation by default, advanced users that want to free space may do so by uninstalling...

For progress bars, see [p_tqdm](https://github.com/swansonk14/p_tqdm), [tqdm with locks](https://github.com/tqdm/tqdm/issues/407#issuecomment-322932800)

There are two options for progress bars: 1) A single bar where 1 iteration = 1 file (p_tqdm, rich) 2) Multiple bars where 1 bar = 1 file, and 1...

For those who are interested, could you please fill [this form](https://docs.google.com/forms/d/e/1FAIpQLSdw0CifeWnbnXGfsS367DY8FEAwXuv9ybTy_Q80z3KgI2OBJA/viewform?usp=sf_link) by the 29th September? I'm planning to submit on the 30th :) I will soon add the formatted paper...

Hi @igordertigor, yes there's still time. You can fill the form and I'll add you to the list :)

@igordertigor just added you to the list

Hi @sneakers-the-rat ! Thanks for your feedback! I've been wanting to add unit tests for some time (#94). I will prioritize it so they are implemented for the main features...

That would be awesome, thank you!