Thanks for your answer. Great work!
Note: you can subtract 2 spinoramas already in the compare tab (choose CEA2034 with splitted views in the drop down menu)
Great!! Thanks!! Didn't notice that before.
the code is open source. Pull requests are welcome.
There is a python library
src/spinorama
and a template for generating the website in js/css/html:
src/website
The python part is not really designed to add new features but it could be refactored easily to do so. As you would expect there are functions loading various formats, computations on this data (from F3 to Harman score, cea2034 ...) and functions to generates graphs. The execution is distributed across machines/cores via
ray.
I looked with some care the code some month ago and I think it's perfect for what it is right know. It's not a minor effort...
Open feature requests on github if you have ideas of what you would like to be able to do.
I was thinking about a broader approach, like a library with audio functions in general: speakers, electronics, engineering, statistics on blind tests...
It would be nice to have all data about speakers and electronics as pandas datasets to be able to quickly perform some data analysis, like comparing speakers by the slope of the linear model of the in-room response.
It could be the library everybody in the audio community would be using: amateurs, reviewers, teachers, students...
At some point, I don't see the benefit of posting the pictures here on the reviews, and it would be very nice just to embed or link to plots made by this libraries from the data, so all plots would be identical, all reviews would potentially have the same plots... I don't know, I miss the 3D balloons in the 8030 review and it's a bit absurd we can't have them as we have the data.
One can think of thousands of concrete functionals, right? I don't know... generate an audio signal sample (pink noise or whatever is appropriate) for a given frequency response to be able to "hear" the differences... be able to subtract the data to then plot it to be able to get, for instance, a difference of horizontal globe... calculate the SPL of a headphone from the input voltage without having to enter its sensitivity because it's already on the library... Also, there could be Jupyter notebooks of analysis made here on ASR about speakers tests or audibility...
I know, all of this is beyond the capabilities of a single person working in spare time, just some random wishful thoughts. Moreover, I'm far from being an expert not even an educated person in audio, so I'm not in position to say what would be useful or make sense and what not.