AUDIBLE FREQUENCY RANGES 



619 



age amplifier and measuring the sound pressure in the listening room 

 with the regular system microphone and amplifier, it is necessary to 

 assume that the microphone behaved identically in the microphone 

 room. The two rooms are similar and the assumption was thought 

 justified. The power output capacity of the system was estimated at 

 one-half watt peak sound power with 10 per cent distortion products. 

 In addition to the speech circuits, Fig. 1 shows an indicating-lamp 

 circuit. Placed before the loud speaker was a small box bearing the 

 letters A and B on its translucent face. As the filters were thrown in or 

 out the illumination was changed from one letter to the other, the letter 

 corresponding to "filter-in" being determined by the reversing switch. 

 The signal lights beside the microphone were lighted whenever the 

 circuit was closed through. An order-wire circuit (not shown) was 

 used for signalling and intercommunication. 



500 1,000 



FREQUENCY IN CYCLES PER SECOND 



5,000 10.000 20.000 



Fig. 3 — Reproduction-ratio characteristic of complete system. 



Testing Methods 



The "A-B Test" method was used in determining the filter cut-offs 

 producing just perceptible changes in the sounds. The observers 

 listened to two conditions, A and B, one filtered and one unfiltered, and 

 judged which condition was filtered. If the observer obtained a score 

 of 100 per cent correct judgments in a large number of trials the filter 

 was absolutely detectable. A score of 50 per cent correct judgments 

 indicated an undetectable filter, because the observer, if guessing, 

 guessed right and wrong an equal number of times. 



Nine members of an articulation testing crew and two young engi- 

 neers made up the regular observing personnel, though other observers 

 were secured when possible. The actual number participating in the 

 tests varied from nine to fourteen. All were known to have normal 

 hearing, but the predominantly youthful makeup of the crew probably 

 made the crew's average sensitivity for very high frequencies somewhat 

 greater than the general average. The observers were frequently 

 shifted about to insure average results, since the sound field was not 

 absolutely uniform. 



