Single-frequency Sudden Noise Annoyance Subjective Experiments

With the rapid development of high-speed rail transit, it has brought convenience to people's travelling, at the same time also caused trouble to the residents around rail transit. The noise of the high-speed rail transit disturbs people's normal life, Formulating high-speed rail transit noise standard can solve this problem and overcome the bottleneck of the high-speed rail transit and facilitate us further development.Standard can't be established without the solid foundation of theory and practice. Therefore it must make an in-depth analysis of the effects of this kind of noise disturbing people, and build the database of annoyance. This paper aims at exploring the effects of such noise on people.The characteristics of high-speed rail traffic noise is rising and falling in short time, so we can call it sudden noise, the experiment is using Artemis 3.0 to make different time-varying stimulus, all the stimulus can be divided into two types, rising form and falling form. All the factors of the stimulus are based on the characteristics of the high-speed rail transit according to different onset rate, offset rate and max sound level under different speeds. The psychoacoustic experiment was carried out in a sound proofing room, Using the Semantic differential (SD) method, the experiment scored the annoyance upon the different stimuli. The effects of different signals upon people can be determined according to the scores rated by the subjects. By using the analysis software SPSS to analyze the data of experiment systematically, the experiment has reached some preliminary conclusions, and established a certain kind of annoyance prediction model. Main conclusions are as follows:1) When the rising signal is the same to onset rate, the results show that the annoyance level is proportional to the max SPL, Following the same logic in the case of the dropping signal, the annoyance level is proportional to the max SPL too;2) When the rising signal is the same to max SPL, the annoyance is inversely proportional to the onset rate, and when the onset rate is less than or equal to 15 dB/s, the annoyance caused by the sudden sound is stronger than the onset rate of more than 15 dB/s signals'annoyance, dropping signal following the same result with the rising signals in the same condition.3) By comparing with the annoyance caused by rising signals and the dropping signals, It can be found the annoyance caused by rising signals is stronger than dropping signals. Therefore with regard to a sound with both the rising and dropping stages, as long as the rising stage does not annoy people, it's guarantied that the whole sound does not annoy people.4) In the built-in condition with fixed rising time during the fourth experiment, the analyzing results show that the annoyance level is proportional to the onset rate, mainly because in the certain time and the bigger onset rate the higher max SPL can reach, so the stronger the influence on people.5) Through multiple linear regression analysis method on the rising signals, the annoyance prediction model is established. This paper mainly introduced the experiment process data and its final conclusions. Due to some limit of some subjective and objective factors, the conclusions need to be further tested. It's hoped that this research will serve as a reference for further research.