Fabian-Robert Stöter

> From the handbook, it takes 3 or 4 parameters. [SDR, SIR, SNR, SAR] = bss crit(s target, e interf, (e noise), e artif) you need to compare with [SDR,SIR,SAR,perm]=bss_eval_sources(se,s)

> In my case, say I have a clean file, "clean.wav"; noise, "noise.wav", (and the mixed.wav) and the denoised file, "denoised.wav". They are with the same length. How to use...

> Do we have the counterpart for bss_crit()? see https://github.com/craffel/mir_eval/blob/master/mir_eval/separation.py#L803 > What's the difference between these two? bss_eval_sources() and bss_crit()? Please read [here](https://hal.inria.fr/inria-00564760/PDF/PI-1706.pdf) > And why you say that mir-eval...

What about also supporting the [multichannel measures](http://bass-db.gforge.inria.fr/bss_eval/bss_eval_images.m)?

Back to the state of bss_eval, I want trigger some discussion about what we could contribute to mir_eval and what would be the possible use cases, for me that would...

@craffel > Usually we do Matlab (or other pre-exising implementation) comparison in the PR, and don't include it directly in the repository. All of the json data is actually for...

@Js-Mim yes, actually it is already finished, but we are waiting for #199 to be ready, so that we get the interface right. Also @ErikJohnsonCU is still looking if we...

@Js-Mim thanks. yeah, maybe we add a pure python version here and highly optimised version in [dsdtools](https://github.com/faroit/dsdtools) where we already know that processing these 5 min tracks take ages to...

@justinsalamon we discusses this among some people in the audio separation community and we decided that it makes more sense to develop future bsseval modules outside of mir_eval since the...

> Hmm, I'm not sure I agree with that statement. I've read several papers (recently) using mir_eval to evaluate BSS systems (rather than the matlab based bss_eval), and I know...