PAQM estimates audible quality degradation by comparing specific loudness across critical bands.

Evidence

A relatively simple auditory model is presented in Figure 17.5; this implements the principle of audio quality evaluation shown in Figure 17.4 (Beerends and Stemerdink, 1992). The model is based on computing the specific loudness for each critical band using time-windowing, DFTtransfer, and frequency warping, in a similar manner to the Matlab script shown in Section 13.1. The specific loudness spectra are compared, and a time-dependent estimate of the audible difference is obtained, which is then averaged to obtain the final estimate of quality degradation. There are many parameters in the model that are selected to match the estimate with results obtained in MOS listening tests. The measure is called the perceptual audio quality measure (PAQM). — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2370

Categorization

• DEFINE: Explains what PAQM is and how it derives from critical-band specific loudness.