I didn't do any technical analysis, but to me RNN sounded as if they had just cut all noise between the narrators speech, and played all sound during their words, giving you a sharp amplitude difference and "stabbing" sensation.
At least that's my take on why Speex is subjectively more pleasant, with which I agree.
Re:RNNoise: Learning Noise Suppression
by zimpenfish on Friday September 29, 2017 @08:56AM
(#15364777)
That's it, yeah, a kind of gated ramp effect at the beginning and end of words.
That means if they had access to the gate we can change that to our taste. Hope that is an option in the function.
To me it sounds like the kind of flange-y wafty MP3 glitches you used to get. At 0dB on the babble sample, it's painful to listen to whilst Speex is perfectly fine.
For me, Speex wins on all samples.