The author investigated the temporary threshold shift(TTS) and its recovery process after the short duration of stimulation of the loud noise and the loud pure tone given to Koreans with normal hearing ability in order to clarify the actual condition of TTS and to set up the preventive indices for noise-induced hearing loss. The subjects were 69 persons(137 ears) whose average hearing was less than 15dB at frequencies from 250Hz to 6,000Hz and less than 10dB at 4,000Hz. The stimulated sounds were 4,000Hz pure tone, narrow band noise of frequency with 4,000Hz at its center, and white noise, and stimulated intensity was 90dB above the threshold in each subject. The mode of stimulation was continuous and intermittent. The continuous tone was given with white noise by 30-second stimulation and 30-second rest 20 times repeatedly for 20 minutes. This investigation was conducted with the conventional and the Bekesy audiometer(Reference to ISO). The following results of investigation were obtained. The average TTS by each stimulated sound with the conventional audiometry was 21.54±0.91dB in 4,000Hz pure tone, 24.42±0.94dB in the narrow band noise, of frequency with 4,000Hz at its center and 22.46±0.83dB in the white noise but there was no statistical significance in each stimulated sound. In case of using the Bekesy audiometer, the average TTS of each stimulated sound was 34.45±1.50dB, 33.03±1.29dB, and 31.67±1.21dB respectively but there was no difference either. By the mode of sound stimulation with the conventional audiometry, the average TTS was 22.46±0.83dB in the continuous white noise and 20.26±0.89dB in the intermittent white noise, and there was a statistical significance between them. With the Bekesy audiometry, the average TTS of the continuous white noise was 31.67±1.21dB and the intermittent white noise was 26.85±1.07dB, and a statistical significance was also shown. All subjects recovered half quantity of TTS at 30 seconds after each sound stimulation, but as time passed, they made a slow and a little recovery, and the process of recovery was exponential in time, which means that if one plots the TTS against the logarisms of the recovery time, a straight line will result. In general, there was a tendency to make a longer recovery in the larger TTS. About 3.8% of ears tested had over 41dB of TTS in the conventional audiometry and 23.9% in the Bekesy one. This TTS over 41dB took a longer recovery time excessively, so there was a doubt that it was susceptible to noise-induced hearing loss. The incidence of ears which took over ten-minute recovery time was 2.7% in the conventional audiometry and 5.8% in the Bekesy one, and a doubt was raised about the possibility of causing noise-induced hearing loss. The rate of positive recruitment phenomenon was 24.6% in ears with 10-minute recovery time and 5.9% in the ears beyond ten-minute. The difference of TTS between both ears was shown in 16.8% with the conventional audiometry and 36.6% with the Bekesy one. In general, TTS was larger and its recovery time was longer in case of using the Bekesy audiometry than the conventional one.