Quote:
Originally Posted by Ghitulescu
The purpose of these tests is to assess the digital loss due to lossy compressions using psychoacoustics. It's no use to compare two different WAVs but consecutive generations of the same original.
|
Sure, most objective metrics are "full reference" metrics, i.e. they compare the compressed/modified audio against the original audio.
If you want to compare the quality of two Wave files using such an objective metric, you'd use the metric to compare each Wave file against the "perfect quality" source file individually.
Then you can compare the results of the two Wave files. The Wave file that got the "better" score obviously has the higher quality (with respect to the metric used).