Narrow frequency bands allow removal of correlation and silence flagging

Evidence

These can be normalized by applying an appropriate scale factor to each band, which effectively removes correlation, and the narrower the bands are the more effective this removal can be. It also allows the coder to flag bands that contain no energy at all and therefore no information needs to be transmitted. — David Howard, Jamie Angus, coustics and Psychoacoustics, 2605

Categorization

Category: Definitional Content

Related

• Narrower frequency bands make audio signals less correlated and easier to quantize (similarity 0.89)
• Narrowband noise needs bandwidth and center frequency descriptors (similarity 0.84)
• Narrowband noise requires bandwidth and center frequency parameters beyond power. (similarity 0.83)
• Frequency-domain coding using masking avoids transmitting masked sound information (similarity 0.81)
• Hearing segregates sounds by frequency unless they fall within a critical band (similarity 0.81)
• Narrowband noises contain fewer frequencies and show greater amplitude fluctuations than broadband noise. (similarity 0.81)
• Any sound narrower than a critical band produces identical loudness regardless of its width. (similarity 0.80)
• Digital noise can be shaped to hide where hearing is least sensitive (similarity 0.80)
• Sound spectrum analysis depends on banks of filters to decompose signals (similarity 0.80)
• Filtering and resonance are more naturally understood in the frequency domain. (similarity 0.79)
• Noise is defined relationally against the signal rather than acoustically (similarity 0.79)
• Noise fluctuations within narrow critical bands raise the masking index (similarity 0.79)
• Broadband noise yields better intensity discrimination than pure tones by engaging multiple frequency regions (similarity 0.78)
• Narrowband noises fluctuate more than wideband stimuli, making them useful for studying fluctuation coding (similarity 0.78)
• Active noise cancellation dominates below 1 kHz while passive isolation takes over above it (similarity 0.78)