【作者】 张昌佳；

【导师】 吴硕贤；

【作者基本信息】 华南理工大学 ， 建筑技术科学， 2010， 博士

【摘要】 体育馆声学设计的主要目的是为了保证语言清晰,而影响室内语言清晰度的两大因素为信噪比和混响。所以,要研究体育馆的声学问题,必然面临几个基本问题:用什么指标来评价体育馆的语言清晰度?体育馆信噪比对语言清晰度的影响如何?为满足语言清晰度,体育馆的混响指标的限值是多少?为解决这些问题,本文从测量出发,做了三方面的实验工作:首先测量了体育馆的声场,包括10座体育馆的背景噪声测量以及4座体育馆的脉冲响应测量。背景噪声测量考虑了体育馆在空场、满场观众等不同状态下的背景噪声的差别;体育馆脉冲响应测量则是采用01dB系统得到了馆内各测点在使用扩声系统时的脉冲响应。其次是体育馆声场计算机仿真实验。在总结出12种体型体量的典型体育馆基础上,采用ODEON软件进行声场仿真。仿真采用了2种扩声方案和3种吸声方案,得到了各受声点的声学参数和双耳脉冲响应。最后进行了体育馆汉语语言清晰度主观评价实验,包括2座体育馆9个听音位置的现场听音实验以及典型体育馆32个听音位置的运用可听化技术进行的听音实验。通过对前述实验数据的总结和分析,针对与体育馆语言清晰度相关的客观声学指标进行了如下三方面的探讨并得出了相应结论。第一是对体育馆语言清晰度客观指标的讨论。本文推荐使用语言传输指数STI作为体育馆语言清晰度客观评价指标,因为体育馆汉语语言清晰度主观评价实验表明,与其他语言清晰度客观指标相比,如%ALcons、D、C50等,STI对语言清晰度得分高低的反映表现最好。另外,主观评价实验还表明,为保证体育馆语言清晰,STI值应大于0.5。第二是体育馆信噪比的考察。首先测量表明体育馆背景噪声在时域和频域上都有其自身的特点。其次,根据背景噪声声压级与扩声系统工作时声压级大小可知体育馆信噪比基本都能达到15dB,这时背景噪声对语言清晰度的影响较小,主观评价实验验证了这一点;信噪比为10dB时,主观实验表明语言清晰度受一定程度的影响。第三是对体育馆混响指标的探讨。体育馆混响指标主要有混响时间与早期衰变时间,主观评价实验表明:与T30相比,EDT与语言清晰度的关系更密切,典型体育馆模拟以及基于体育馆声能衰变特点的理论分析也表明EDT与STI的相关性更高。体育馆混响指标限值较难界定,本文提炼的典型体育馆声场模拟结果得到的T30（EDT）与清晰度得分的三次多项式回归曲线显示,T₃₀小于3.5s（EDT<3s）,汉语语言清晰度得分大于75%。另外,行业标准对体育馆混响时间的限制比较严格,要达到其要求,需要屋顶和墙面都布置强吸声材料;同时,规范推荐的伊林公式计算混响时间的准确性还值得进一步研究。更多还原

【Abstract】 Speech intelligibility is the main aim of acoustical design in a sports hall and two factors that influence speech intelligibility are signal to noise ratio and reverberation. Above is the foundation of the writing and arrangement of this paper, which brings about the following questions: which objective index should be used to evaluate the speech intelligibility of a sports hall? How does the background noise influence speech intelligibility of a sports hall? How long the RT (or EDT) should be in order to obtain good speech intelligibility of a sports hall and how to control the reverberation in a sports hall?In order to answer these questions, experiment work in three fields has been done starting from measurements of sound field:First, the sound field of several sports halls is measured, including measurements of the background noise 10 sports halls and impulse response in 4 sports halls. The former measurement takes into account the differences between background noise when the sports hall is unoccupied, occupied and in other state, and in the latter measurement, the 01db system is utilized to obtain the impulse response of each receiver when using the loudspeaker system.Second, the sound field of some typical sports halls has been simulated. Based on the summarization of 12 typical sports halls with different shape and size, ODOEN is used for sound field stimulation. Two types of loudspeaker systems and three types of absorption material arrangements are used in the stimulation, and the acoustic parameters and binaural impulse response of each receiver is obtained.Third, the subjective evaluation test of speech intelligibility of sports halls has been conducted, including test of 9 listening positions in two sports halls and auralization test of 32 listening positions of typical sports halls.On the basis of the summarization and analysis of the experiment data, the following disccussions is made and conclusions drawn according to the acoustic parameters related to speech intelligibility.Firstly, the index of speech intelligibility of sports halls has been discussed. This paper suggest STI be used for the evaluation, because the subjective evaluation test of speech intelligibility illustrates that compared with other parameters such as %ALcons, D and C50, STI can better evaluate speech intelligibility. In addition, subject evaluation test shows that STI should be more than 0.5.Secondly, signal to noise ratio in a sports hall has been investigated. Investigation finds significant difference between background noises when the sports hall is unoccupied and occupied. For example, the background noise SPL is about 76 dB when audience exchange casually in an occupied sports hall, which is 35dB more than when the hall is unoccupied. The signal to noise ratio of sports halls generally can reach 15dB, and the background noise has little influence on the speech intelligibility under such circumstance, which subjective tests have testified. When the signal to noise ratio is 10dB, subjective tests demonstrate that the speech intelligibility is influence to some extent.Third, the reverberation index in sports halls has been discussed. RT and EDT are two main index of reverberation. Subjective evaluation tests indicate EDT is more closely related to speech intelligibility than T30, and the simulation results of typical sports halls as well as the theoretical analysis of energy decay indicates that the relationship between EDT and STI is closer. Therefore, it is more suitable to use EDT as reverberation index for acoustic design of sports halls. It’s difficult to determine how long the reverberation time should be in a sports hall. In this paper, the simulation results of sound field of typical sports shows: if T30<3.5s(EDT<3s), the Chinese speech intelligibility score can be more than 75%. In order to reach the reverberation limited value demanded by standard, the ceiling and walls should be filled with absorptive material, this is rigrid for acoustical design; the inccuration of reverberation time calculated by Erying formula which suggested by stardand should be discussed in the next work.更多还原