THE SOUNDS OF Sl'EliCll 



001 



spokiMi in .m ordinary tone of voiir, with tlu- spciikiT's mouth al)out 

 thri-e inrlii's from the transmitter. A key was pri-ssrd l)y the s|)i'aker 

 just before tlie sound was spoken, lliis releasing a shutter plaie<l lie- 

 fore a rotating him drum on whiih the record from the oscillograph 

 vibrator was traced. The film drum was some ")2 inches in circumfer- 

 ence, and there was mounted on it a length of Kastman super-speed 



Fi^. 1(1 — Caliiilatecl ami hIi-mimiI w.ivc iDriiis, as rccorfled by the apparatus 



film with which records coukl be made at a peripheral speed of about 

 20 feet per second. Thus each hundredth of a second corresponds to 

 two inches or more in the time scale on the film. Besides opening the 

 shutter, the key released a mechanism which swung the oscillograph 

 vibrator through an arc during the progress of the record, thus tracing a 

 helical record on the film. By this means records up to 200 inches 

 in length, or for nearly one second of duration were taken. The 

 average length of the wave trains recorded was less than 0.5 second; 

 thus it was possible to graph the pressure wave of the whole speech 

 .sound from beginning to end. Immediately following the recording 

 of the speech sound a timing wave of 1,000 cycle alternating current, 

 taken from a standard oscillator, was recorded on the film at one side 

 of the speech record, without disturbing the speed adjustment of 



