MOS (Mean Opinion Score) is the foundational measure of perceived audio quality: a panel of listeners rates samples on a five-point scale (1 bad, 5 excellent) and the scores are averaged. It is subjective by design and therefore the ground truth that every objective metric — PESQ, POLQA, ViSQOL — is built to predict. MOS appears throughout codec evaluation, VoIP monitoring, and standards work, often via specific test methodologies like ACR (absolute category rating). Its weaknesses are cost, time, and variability — results depend on the listeners, the content, and the test conditions — which is exactly why automated estimators exist. When you see a single quality number for a codec or a call, it is usually a MOS, measured or predicted.
