The Masking of Speech by Sine Waves, Square Waves, and Regular and Randomized Pulses

The ability of three standard wave forms to mask speech, by raising its threshold of perceptibility, was measured as a function of the intensity and frequency of the masking signal. Depending on the other parameters of the interfering signal, optimal masking is produced when the fundamental frequency lies in the range 100 to 500 c.p.s. Intense sine waves mask best when their frequency is about 300 c.p.s., whereas for weaker sine waves the most effective frequency is about 500 c.p.s. Square waves are less critical as to frequency and intensity; fundamental frequencies between about 80 and 400 c.p.s. mask with approximately equal effectiveness. Pulses of 10‐microsecond duration are most effective at a prf of about 200 p.p.s. When the time interval between successive pulses is made irregular (random interval‐modulation), their ability to mask speech is dramatically increased. The increase in masking is a function of the original prf, the rate of modulation (determined by the spectrum of the modulating voltage), and the range of modulation. By a proper adjustment of the parameters of its random interval‐modulation the train of pulses can be made to sound like a band of white noise.