David van Leeuwen

Hi, this would be Python 3.7.3, Linux 4.19.0-6-amd64 #1 SMP Debian 4.19.67-2+deb10u1 (2019-09-20) x86_64 GNU/Linux. I am now trying the same on a 128GB mem sibling computer. I followed the...

Typically I never really execute the do-it-all scripts as I tend to store source databases somewhat differently from what the do-it-all script write had in mind. Also, the do-it-all scripts...

No, I didn't. The instructions about `recreate.sh` suggested that that was optional ("if needed"). I did check the consistency, the script found some 20 differences or so w.r.t. the expected...

Sure, I understand that is is hard to reproduce the problem from the description, and the databases are quite large, but is there a way to get kaldi_writer.save() a bit...

OK, just for the sake of trying, I re-did `cd megs; python scripts/merge_and_subset.py data/download data/full`, and then `cd ..; python local/prepare_data.py megs/data/full/ data`, and this still bails out after 261...

Thanks! This seems to work for me. I use an event `shutdown_event` to consecutively signal - stop accepting new connections - all connections are closed For future reference, the code...

This looks like a numerical instability, indeed, I'd say something which this the output of `cov()` should be positive definite. So, in your case you could look at the `y...

Hello, I am not sure. Somebody closed that PR, it wasn't me. Usually a PR has a bit more explanation on what is going on, and there should be some...

This would be currently out of the scope for this package, maybe you can deal with incidental cases in a proprocessing step. I didn't read the full article, though, to...

Cool!