Masking of sound

Masking of sound

Interference with the audibility of a sound caused by the presence of another sound. More specifically, the number of decibels (dB) by which the intensity level of a sound (signal) must be raised above its threshold of audibility, to be heard in the presence of a second sound (masker), is called the masking produced by the masker on the signal. The masker and the signal may be identical or may differ in frequency, complexity, or time.

When both the masker and signal are pure tones and the tonal signal and masker have the same frequency, a very low level masker is required to mask the signal, indicating significant masking. As the difference in the frequency between the signal and masker increases, the signal is easier to detect, requiring a higher-level masker to mask the signal. Results from these psychophysical tuning curve measures of masking agree very well with data obtained from single auditory neurons in the auditory periphery, suggesting that tonal masking is mediated by the activity of these peripheral neurons.

The most widely studied complex masking sound is random noise which has energy at all frequencies and is said to be flat if the level for each 1-Hz bandwidth of the noise is the same. When random-flat noise is used to mask a pure tone, only a narrow frequency band (critical band) of the noise centered at the tonal frequency causes masking. If the bandwidth of the noise is narrower than this critical bandwidth, the tone's intensity can be lowered before the tone is masked. If the bandwidth is wider than this critical bandwidth, further widening of the bandwidth causes no changes in the detectability of the signal. The width of the critical band increases proportionally to its center frequency, that is, to the signal frequency. When noise masks speech, either the detectability of speech or speech intelligibility can be measured. The level for speech intelligibility is about 10–14 dB higher than for speech detectability.

Masking can occur when the signal either precedes or follows the masker in time. In backward masking the signal precedes the masker, while in forward masking the signal follows the masker. The physiological basis for this effect, as well as its implications for auditory processing of complex stimuli, is of great interest to the auditory scientist. See Acoustic noise, Sound

References in periodicals archive ?
The "soundscape" can be more complex than existing stereo or mono situations permit because QSound facilitates less crowding or masking of sounds.