A simple high-pass filter can replace elaborate loudness models

Evidence

Section 10.2 introduced a simple loudness model that accurately estimates the loudness perception evoked by many relatively simple signals. Such models have been found to be unable to derive accurate loudness estimates for spectrally complex and time-variant signals. This shortcoming has motivated the design of more advanced loudness models (Florentine et al., 2005), of which two approaches are briefly touched upon next. The first one (Zwicker, 1977) processes the signal in the time domain and produces a continuous signal representing the loudness as a function of time. This approach opens up the possibility of predicting the loudness perception evoked by a time-variant signal based on the peak values in the model output. The second model (Glasberg and Moore, 2002) divides the signal into overlapping time frames and extracts short-term loudness values from each time frame. The model also emulates the temporal integration of loudness between adjacent time frames, and consequently, the model provides accurate estimates for time-variant signals as well, although this estimate is derived by simply averaging across the short-term loudness values. Despite the improved accuracy in predicting perceived loudness of complex signals, the different models cannot yet fully explain loudness perception. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 1659 (#complicates, via linked claim)

In order to have a better, yet relatively simple, measure of the perceived level of sound, the concept of sound level is defined as a measure that is weighted with frequency so that the level roughly approximates the frequency sensitivity of hearing. Four different weighting curves are defined: A, B, C, and D, as shown in Figure 9.3. The A-weighted sound level is commonly used in noise measurements to characterize the perceived level and the risk of hearing loss. The A weighting slightly emphasizes the levels at mid frequencies and attenuates them at low and high frequencies. When the weighting curve is compared with the inverted equal-loudness curves, it is easy to see that the A weighting is just a very rough estimate. Technical simplicity and extensive use in practice are its advantages. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 988 (#expands, via linked claim)

Categorization

• XREF: connects to: ISO standardized loudness-based noise metrics because A-weighting proved inadequate; Loudness-calculating procedures outperform simple SPL for practical noise reduction; Noise reduction claims should use loudness ratios to stay credible — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 1782 (+3 more)

Related

• ISO standardized loudness-based noise metrics because A-weighting proved inadequate (similarity 0.87)
• Two competing loudness models predict time-variant loudness by peak tracking or frame averaging (similarity 0.87)
• Loudness-calculating procedures outperform simple SPL for practical noise reduction (similarity 0.87)
• Perceived loudness of time-varying sounds tracks near-peak levels, not averages (similarity 0.87)
• Sound level meters trade average level against tracking speed via time response. (similarity 0.86)
• A-weighting is recommended for all sound levels despite fitting low levels best (similarity 0.86)
• A-weighted decibels are a rough practical approximation of perceived loudness (similarity 0.86)
• Loudness meters use frequency weightings as a compromise because ear response varies with level (similarity 0.85)
• A-weighted sound level approximates perceived loudness only crudely, yet remains the noise-measurement standard. (similarity 0.85)
• Loudness scales with frequency squared, requiring large amplitude at low frequencies (similarity 0.85)
• A-weighting can miss dominant low-frequency energy in broadband sounds (similarity 0.84)
• A-weighted dB values misrepresent perceived loudness for complex sounds. (similarity 0.83)
• No single frequency-weighting can accurately meter loudness at all levels (similarity 0.83)
• Static loudness models fail on time-variant sounds, motivating frame-based temporal integration models (similarity 0.82)
• Loudness measures can reveal real improvements A-weighting misses entirely (similarity 0.82)

Broader Connections

• Two competing loudness models predict time-variant loudness by peak tracking or frame averaging #complicates (0.75) — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 1659
• A-weighted sound pressure level misrepresents the loudness of broadband noises #corroborates (0.75)
• A-weighted decibels are a rough practical approximation of perceived loudness #expands (0.72) — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 988