Synthetic noises deviate from Gaussian masking when line counts or phases are poorly chosen

Evidence

Gaussian noise can be simulated by a large number of low-level tonal components with random phase. Such approximations have to be adopted with care when used as maskers, because of the non-Gaussian amplitude distribution produced in the case of too few lines and/or non-arbitrary phase conditions. The same holds for noises produced by “statistically” chosen sequences of “0” sand “1” s. Such noises often produce masking effects different from those produced by Gaussian noise because of the different amplitude distribution and the different temporal and spectral pattern within narrowfrequency bands. Such patterns are important for the excitation level versus critical-band rate versus time pattern, the basis of masking. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 616

Categorization

Category: Causal Relationships

• DEFINE: Clarifies that amplitude distribution and narrowband temporal-spectral pattern, not just spectral envelope, define how a noise masks.
• QUESTION: Worth chasing: how large must the tonal-component count be for a random-phase approximation to behave as Gaussian?

Related

• Simulated Gaussian noise with too few components gives atypical masking results (similarity 0.93)
• Masked threshold rises once noise bandwidth exceeds the critical bandwidth (similarity 0.82)
• Tone-on-tone masking thresholds are measured via narrow-band noise maskers instead (similarity 0.82)
• Masking relates closely to excitation level patterns. (similarity 0.81)
• Masking reduces to a comparison of excitation-level patterns over frequency and time (similarity 0.81)
• Uniform masking noise mirrors white-noise masking and gives flat thresholds (similarity 0.81)
• White noise masking of tones is linear but frequency-dependent above 500 Hz (similarity 0.81)
• Noise bandwidth below the critical band does not raise masked threshold. (similarity 0.80)
• Five tones replicate narrow-band noise masking while single tones distort it (similarity 0.80)
• Both spectral sparsity and lack of masker fluctuations contribute to release from masking. (similarity 0.80)
• Masking pattern shapes differ by center frequency even at equal masker level (similarity 0.80)
• Uniform masking noise can substitute for threshold-in-quiet across all frequencies. (similarity 0.80)
• Masking pattern curves are scale-invariant and predictable by horizontal shifts (similarity 0.80)
• Masking by bandlimited noise depends only on masker energy within the critical band. (similarity 0.80)
• Pulsation patterns can replicate masking patterns to under 1 dB accuracy (similarity 0.80)