Hello everyone!
I am Maris, a PhD student at ETH Zurich. I was invited by Prof. Cohen after a discussion we had at a conference in Italy recently.
I’d like to contribute to the library, specifically in similarity metrics and possibly new neural network architectures since that’s my area of expertise.
I’m happy to discuss on what contribution can I make.
Kind regards,
Hi Maris, welcome!
You can write in the ‘introduce yourself’ discussion (where you can also get to know other members of the forum).
Also, I’m not a prof… not yet 
Welcome @marisbasha!
So sorry I didn’t see this before!
Like @yardenc said you can definitely feel free to (re)introduce yourself on that thread if you’d like!
I’d like to contribute to the library, specifically in similarity metrics and possibly new neural network architectures since that’s my area of expertise.
Sounds great! @yardenc did tell me you discussed similarity and some ideas about neural networks at the European Songbird conference
re: similarity, if you want to raise an issue proposing your idea on the repo for the VocalPy package (not to be confused with this community of the same name 
) please feel free!
It would be great to have more code in Python to get at this.
@therese.koch and I took a stab at a SAP port to Python last year at the SciPy conference and it’s been lying dormant ever since 
I agree with you that a consistent “interface” for measuring similarity would be helpful, e.g. a class named SimilarityMeasurement with a measure method, or something like that.
One issue might be that people tend to measure different things but still call it similarity, e.g. Kershenbaum’s work on using info theory to measure similarity of sequences
https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.12433
So we might want to have an AudioSimilarity class and a SequenceSimlarity class?
Happy to continue discussing this high-level idea here