search

aliutkus / speechmetrics

A wrapper around speech quality metrics MOSNet, BSSEval, STOI, PESQ, SRMR, SISDR
MIT License
894 stars 153 forks source link

How to comprehend output? #12

Open dmumtaz opened 4 years ago

dmumtaz commented 4 years ago

Hi First, the metric is super cool, it saved me from downloading each of the metrics separately. Thanks! Also I wanted to know how to comprehend the output. It would be great if you add that to the Readme file. Here is the output from two of the files in your dataset, could you elaborate on the results, as in what does high positive or negative value or close to zero mean?

reference = 'data/m2_script1_produced.wav' ditorted = 'data/m2_script1_clean.wav'

{'mosnet': array([[5.0981326]], dtype=float32), 'srmr': 4.653473083972128} {'sdr': array([[-0.39609285]]), 'isr': array([[0.24738725]]), 'sar': array([[-0.37060632]]), ' pesq': 4.354660987854004, 'sisdr': -14.740691053217517, 'stoi': 0.9718856108717927}

aliutkus commented 4 years ago

hi @dmumtaz, thanks for the encouragements. I think that each metric reference should be checked to have a better understanding, but maybe I could at least update the docs to specify the unit of each metric, and whether higher or lower is better. you would like that?

dmumtaz commented 4 years ago

Yes, that would be great.

aliutkus commented 4 years ago

done