Shaping quantization noise to sit 13 dB below the signal renders it inaudible

Evidence

In order to make the reduction of information flow inaudible, it is best to use the characteristics of the human ear as an information receiver, and reduce the information flow for those events that are already inaudible. The effect of masking is a typical example. It can be introduced into the processing system in such a way that those parts masked by loud sounds are not transmitted to the receiver. — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 2020 (#exemplifies, via linked claim)

Although the quantizer will pick the output level that causes minimum error, there will on average always be some error. This error adds noise and distortion to the signal and is often referred to as quantization noise. Ideally this noise should be random and often dither is added to ensure this. The effect of the error is to reduce the signal to noise ratio of the audio signal. If a binary word of Nbits bits is used to encode the audio signal, then the maximum signal to noise ratio is given by: For a 16-bit word (the CD standard), this gives a maximum signal to noise ratio of 96 dB. — David Howard, Jamie Angus, coustics and Psychoacoustics, 2651 (#qualifies, via linked claim)

Categorization

• XREF: connects to: Conventional distortion measurements include inaudible masked components; Clean amplifier clipping is inaudible until waveforms lose over half their amplitude — , Toole, Floyd E - Sound reproduction-…, 1792; Quantization always injects noise that lowers audio signal-to-noise ratio — David Howard, Jamie Angus, coustics and Psychoacoustics, 2651 (+4 more)

Broader Connections

• Quantization always injects noise that lowers audio signal-to-noise ratio #qualifies (0.80) — David Howard, Jamie Angus, coustics and Psychoacoustics, 2651
• Audio compression exploits masking to discard inaudible signal parts #exemplifies (0.80) — Professor Dr.-Ing. Hugo Fastl, Professor Dr.-Ing. Eberhard, Psychoacoustics- Facts and Models, 2020
• Uniform quantization noise requires about 60 dB signal-to-noise to stay masked #complicates (0.78)
• Loud-signal masking of weak signals enables bit-saving adaptive quantizers #expands (0.72) — David Howard, Jamie Angus, coustics and Psychoacoustics, 2657