Predicting subjective listening preference from measured data remains a hard unsolved problem.

Evidence

The problem with subjective evaluation is that the required experiments are typically very laborious and time consuming. The experiments should be carried out in conditions that correspond to real usage of the sounds or related products. Thus, it would be much easier and less expensive to use objective criteria or models for sound quality. Ideally, a computational model provides an estimate of quality from signal analysis which correlates well with subjective data. The development of such a model (Figure 17.1) is typically based on proper subjective listening experiments from which data are collected and analysed statistically. Using various techniques, the factors and features of sounds that best explain the subjective behaviour are sought. The computational model can be a simple linear regression model or a more complex non-linear model, such as a neural network trained to map the feature parameters to quality indices. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2332 (#expands, via linked claim)

Categorization

• XREF: connects to: Subjective experience dictates which measurements matter in audio research; Computational sound quality models are trained on subjective listening data — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2332; Measurement research aims to map objective data onto subjective sound quality ratings (+1 more)

Related

• Computational sound quality models are trained on subjective listening data (similarity 0.93)
• Measurement research aims to map objective data onto subjective sound quality ratings (similarity 0.89)
• No universal sound-quality model exists; listener data is the only shared starting point (similarity 0.86)
• Sound quality cannot be assessed except through psychophysical experiments with human subjects (similarity 0.86)
• Subjective experience dictates which measurements matter in audio research (similarity 0.86)
• Objective sound quality models have limited domains of validity and cannot replace subjective testing (similarity 0.85)
• Psychoacoustic knowledge rests entirely on subjective listening tests, not direct measurement (similarity 0.84)
• Perceptual auditory models will eventually replace traditional audio metrics (similarity 0.83)
• Headphone preference ratings can be predicted from objective measurements like loudspeakers’ can. (similarity 0.83)
• Subjective sound quality is authoritative and measurements must explain it (similarity 0.83)
• Industry relies on objective methods because exhaustive listening tests are impractical (similarity 0.83)
• Headphone sound-quality variance comes more from user and content variables than target-curve refinements (similarity 0.82)
• Digital audio growth made understanding human audition essential (similarity 0.82)
• Psychoacoustic annoyance models only work within limited signal sets. (similarity 0.82)
• Sound quality means judged preferability, not category membership (similarity 0.82)

Broader Connections

• Computational sound quality models are trained on subjective listening data #expands (0.85) — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 2332
• Measurement research aims to map objective data onto subjective sound quality ratings #corroborates (0.82)