Spectral contrast coding can be enhanced by attenuating valleys while leaving peaks fixed

Evidence

The new SCE strategy incorporates a new processing stage just before the NofM band selection. The goal of this stage is to enhance spectral contrast by attenuating spectral valleys while keeping spectral peaks constant. The amount of spectral contrast enhancement can be controlled by a single parameter termed SCE factor. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1831

2.1 The Signal Processing Method: SCE in NofM Strategies for CIs The baseline or reference speech coding strategy is the advanced combinational encoder (ACE, a description of this strategy can be found in Nogueira et al. 2005 ). The ACE strategy can be summarized in five signal processing blocks: (1) The Fast Fourier Transform (FFT); (2) The envelope detector; (3) The NofM band selection; (4) The loudness growth function (LGF) compression and (5) The channel mapping. The new SCE strategy incorporates a new processing stage just before the NofM band selection. The goal of this stage is to enhance spectral contrast by attenuating spectral valleys while keeping spectral peaks constant. The amount of spectral contrast enhancement can be controlled by a single parameter termed SCE factor. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1832

The ACE strategy can be summarized in five signal processing blocks: (1) The Fast Fourier Transform (FFT); (2) The envelope detector; (3) The NofM band selection; (4) The loudness growth function (LGF) compression and (5) The channel mapping. The new SCE strategy incorporates a new processing stage just before the NofM band selection. The goal of this stage is to enhance spectral contrast by attenuating spectral valleys while keeping spectral peaks constant. The amount of spectral contrast enhancement can be controlled by a single parameter termed SCE factor. — Dijk, Pim van, Physiology, Psychoacoustics and…, 1832

Categorization

• DEFINE: Defines what spectral contrast enhancement (SCE) actually does mechanistically: valley attenuation with peaks held constant, controlled by a single SCE factor parameter.
• corroborated: 5 independent passages support this claim — Dijk, Pim van, Physiology, Psychoacoustics and…, 1831; Dijk, Pim van, Physiology, Psychoacoustics and…, 1832
• XREF: connects to: Separate compression channels for peak and valley reduce spectral contrast; More compression channels reduce spectral contrast and vowel recognition; Over-attenuating spectral valleys may remove speech information CI users need — Dijk, Pim van, Physiology, Psychoacoustics and…, 1823

Broader Connections

• Separate compression channels for peak and valley reduce spectral contrast #corroborates (0.75)
• Over-attenuating spectral valleys may remove speech information CI users need #complicates (0.75) — Dijk, Pim van, Physiology, Psychoacoustics and…, 1823
• More compression channels reduce spectral contrast and vowel recognition #qualifies (0.70)