在台灣，學校之禮堂、體育館、社區活動中心等中小型活動中心，常作為演奏、演唱及演講等多功能的使用。不同的使用行為應搭配不同的建築聲學設計基準，以確保包含迴響時間（RT）、語言清晰度（C50）、音樂清晰度（C80）、聲壓級（SPL）分布、語音清晰度（STI）等聲學指標滿足要求。在進行建築設計時，大多僅採用沙賓迴響時間公式進行確認（Sabine’s equation），此類估算公式雖然簡潔迅速，但是在計算過程中忽略很多細節。另外透過Odeon、Ease等室內聲學模擬軟體雖可得到較準確的分析，卻較為複雜且耗時，在實務設計上較少採用。
本研究的目的，是藉由機器學習（Machine Learning）的方式，提出聲學指標的預測模式，作為中小型活動中心進行室內裝修及設計時的簡易評估工具。首先確認聲學模擬軟體預測與現場實測之再現性，接著透過參數化設計方法生成800個空間樣本做為分析對象，採用Odeon進行模擬分析取得各項聲學指標的預測值，利用機器學習的監督式學習方法（Supervised Learning），透過空間基本的幾何訊息、材料特性、擺放位置等參數進行訓練後得到預測模型。將資料以80%及20%的比例分配給訓練集和測試集，並以測試集的資料樣本進行模型效能的評估，以確認預測模型的適用性。
結果發現，透過GBDT及ANN演算法，在各項聲學指標的預測幾乎皆可達到JND ± 2以內的成效，在C50、C80、STI及聲壓值分布差值的JND，更可達到 ± 1以內。其中迴響時間的預測以GBDT最為準確，相比於傳統之計算公式，其預測能力較好。其他聲學指標則以ANN的預測效果為最佳。透過此方法，可以建立方便迅速並具有精準度的聲學指標預測模型，不須透過建模及聲學模擬軟體即可得知空間中的各項聲學指標。

In Taiwan, small and medium-sized activity centers such as school auditoriums and gymnasiums are common multi-functional spaces that are often used for performances, singing, and speeches. However, most cases are designed using only Sabine’s equation for architectural acoustics. Although that estimation formula is simple and fast, the calculation process ignores many details. Furthermore, while more accurate analysis can be obtained through acoustics simulation software, it is more complicated and time-consuming and thus is rarely used in practical design.
The purpose of this study is to use machine learning to propose a predictive model of acoustic indicators as a simple evaluation tool for the architectural design and interior decoration of multi-functional activity centers. We generated 800 spaces using parametric design, adopting Odeon to obtain acoustic indicators. The machine learning model was trained through basic geometric information, material properties, and other parameters of the space, and the performance of the model was evaluated with the testing set.
We found that through GBDT and ANN algorithms, almost all acoustic indicators could be predicted within JND ± 2, and the JND of C50, C80, STI, and the distribution of SPL could reach within ± 1. GBDT is the most accurate predictor of RT and is superior to traditional calculation formulas. For other acoustic indicators, the performance of the ANN model is the best. Through machine learning methods, we established a convenient, fast, and accurate prediction model and were able to obtain various acoustic indicators of the space without 3D-modeling or simulation software.

(一) 中文文獻

1. 周鼎金編（1995）。建築物理。台北市：裕祥出版社。
2. 台南市里社區活動中心入口網站_線上借用場地。檢自https://tnda.tainan.gov.tw/acthouse/（May. 19, 2020）。
3. 新竹縣政府公開資訊_公有場地租借資料。檢自https://www.hsinchu.gov.tw/cl.aspx?n=461（May. 19, 2020）。
4. 張斐章、張麗秋、黃浩倫（2003）。類神經網路：理論與實務。台北市：台灣東華書局股份有限公司。
5. 曾品緁（2020）。摺板結構金屬擴張網之吸音性能研究。國立成功大學建築研究所碩士論文，台南市。
6. 教育部主管法規查詢系統_國民小學及國民中學設施設備基準（2002）。檢自https://edu.law.moe.gov.tw/LawContent.aspx?id=GL000314（Mar. 07, 2021）。
7. 日本建築學會（2013）。建築設計資料集成。台北市：詹氏書局。
8. 台中市政府服務e平台_場地租借。檢自https://eservices.taichung.gov.tw （May. 19, 2020）。
9. 台北市政府公有場地租用。檢自https://service.gov.taipei/rental （May. 19, 2020）。
10. 彰化縣縣立各級學校校園場地開放使用資訊平台_校園場地租用。檢自http://ishare.chc.edu.tw（May. 19, 2020）。
11. 黃士嘉、林邑撰（2019）。輕鬆學會Google TensorFlow 2.0：人工智慧深度學習實作開發。新北市：博碩文化股份有限公司。

(二) 英文文獻

1. Alpaydin, E. (2010). Introduction to machine learning. MIT press.
2. Arau-Puchades, H. (1988). An improved reverberation formula. Acta Acustica united with Acustica, 65(4), 163-180.
3. Barron, M. (2009). Auditorium acoustics and architectural design. Routledge.
4. Beranek, L. L. (2006). Analysis of Sabine and Eyring equations and their application to concert hall audience and chair absorption. The Journal of the Acoustical Society of America, 120(3), 1399-1410.
5. Bishop, C. M. (2006). Pattern recognition and machine learning. springer.
6. Bork, I. (2000). A comparison of room simulation software-the 2nd round robin on room acoustical computer simulation. Acta Acustica united with Acustica, 86(6), 943-956.
7. Bradley, J. S. (1986). Predictors of speech intelligibility in rooms. The Journal of the Acoustical Society of America, 80(3), 837-845.
8. Breiman, L. (1997). Arcing the edge. Technical Report 486, Statistics Department, University of California at Berkeley.
9. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
10. Cairoli, M. (2018). Architectural customized design for variable acoustics in a multipurpose auditorium. Applied acoustics, 140, 167-177.
11. Cairoli, M. (2020). The architectural acoustic design for a multipurpose auditorium: Le Serre hall in the Villa Erba Convention Center. Applied Acoustics, 173, 107695.
12. Ciregan, D., Meier, U., & Schmidhuber, J. (2012, June). Multi-column deep neural networks for image classification. In 2012 IEEE conference on computer vision and pattern recognition (pp. 3642-3649). IEEE.
13. Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3), 273-297.
14. Cucharero, J., Hänninen, T., & Lokki, T. (2019, September). Influence of sound-absorbing material placement on room acoustical parameters. In Acoustics (Vol. 1, No. 3, pp. 644-660). Multidisciplinary Digital Publishing Institute.
15. Documentation of Scikit-Learn 0.22.2. Retrieved from: https://scikit-learn.org/0.22/index.html (Feb. 25, 2021).
16. D'Orazio, D., Fratoni, G., & Garai, M. (2017). Acoustic design of a multipurpose hall inside a former church. IN_BO. Ricerche e progetti per il territorio, la città e l'architettura, 8(11), 570-582.
17. Duanqi, X., Zheng, W., & Jinjing, C. (1991). Acoustic design of the gymnasium and natatorium of chinese national olympic sports center. Applied Acoustics, 34(4), 267-279.
18. EASE Software. (2009). EASE 4.3 User’s Guide & Tutorial. Germany: ADA (Acoustic Design Ahnert).
19. Eyring, C. F. (1930). Reverberation time in “dead” rooms. The Journal of the Acoustical Society of America, 1(2A), 217-241.
20. Falcon Perez, R. (2018). Machine-learning-based estimation of room acoustic parameters (Master’s thesis, Aalto University, Espoo, Finland).
21. Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of statistics, 1189-1232.
22. Gamper, H., & Tashev, I. J. (2018, September). Blind reverberation time estimation using a convolutional neural network. In 2018 16th International Workshop on Acoustic Signal Enhancement (IWAENC) (pp. 136-140). IEEE.
23. Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. Cambridge: MIT press.
24. Gordon, D. (2010). Multipurpose Spaces. National Clearinghouse for Educational Facilities.
25. Harris, C. M. (1997). Handbook of Noise Control, McGraw Hill
26. Hensen, J. L., & Lamberts, R. (Eds.). (2012). Building performance simulation for design and operation. Routledge.
27. Hinton, G., Deng, L., Yu, D., Dahl, G. E., Mohamed, A. R., Jaitly, N., ... & Kingsbury, B. (2012). Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups. IEEE Signal processing magazine, 29(6), 82-97.
28. Houtgast, T., & Steeneken, H. J. (1985). A review of the MTF concept in room acoustics and its use for estimating speech intelligibility in auditoria. The Journal of the Acoustical Society of America, 77(3), 1069-1077.
29. IEC 60268-16:2011 Sound system equipment — Part 16: Objective rating of speech intelligibility by speech transmission index. International Electrotechnical Commission.
30. ISO 11654:1997 Acoustics — Sound absorbers for use in buildings — Rating of sound absorption. International Organization for Standardization.
31. ISO 3382-1:2009 Acoustics — Measurement of room acoustic parameters — Part 1: Performance spaces. International Organization for Standardization.
32. ISO 354:2003 Acoustics — Measurement of sound absorption in a reverberation room. International Organization for Standardization.
33. Kaarlela-Tuomaala, A., Helenius, R., Keskinen, E., & Hongisto, V. (2009). Effects of acoustic environment on work in private office rooms and open-plan offices–longitudinal study during relocation. Ergonomics, 52(11), 1423-1444.
34. Khemchandani, R., & Chandra, S. (2009). Knowledge based proximal support vector machines. European Journal of Operational Research, 195(3), 914-923.
35. Krokstad, A., Strom, S., & Sørsdal, S. (1968). Calculating the acoustical room response by the use of a ray tracing technique. Journal of Sound and Vibration, 8(1), 118-125.
36. Kulowski, A. (1985). Algorithmic representation of the ray tracing technique. Applied Acoustics, 18(6), 449-469.
37. Kuncheva, L. I., Bezdek, J. C., & Duin, R. P. (2001). Decision templates for multiple classifier fusion: an experimental comparison. Pattern recognition, 34(2), 299-314.
38. Lewers, T. (1993). A combined beam tracing and radiatn exchange computer model of room acoustics. Applied Acoustics, 38(2-4), 161-178.
39. Nannariello, J., & Fricke, F. (1999). The prediction of reverberation time using neural network analysis. Applied Acoustics, 58(3), 305-325.
40. Nannariello, J., & Fricke, F. R. (2001). The use of neural network analysis to predict the acoustic performance of large rooms Part I. Predictions of the parameter G utilizing numerical simulations. Applied Acoustics, 62(8), 917-950.
41. Naylor, G. M. (1993). ODEON—Another hybrid room acoustical model. Applied Acoustics, 38(2-4), 131-143.
42. Odeon Acoustic Software. (2018). Odeon Room Acoustics Software User Manual Version 14. Lyngby, Denmark: ODEON A/S.
43. Passero, C. R. M., & Zannin, P. H. T. (2012). Acoustic evaluation and adjustment of an open-plan office through architectural design and noise control. Applied Ergonomics, 43(6), 1066-1071.
44. Patel, J., Shah, S., Thakkar, P., & Kotecha, K. (2015). Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques. Expert systems with applications, 42(1), 259-268.
45. Rindel, J. H. (2000). The use of computer modeling in room acoustics. Journal of vibroengineering, 3(4), 219-224.
46. Sabine, W. C. (1922). Collected papers on acoustics. Cambridge, MA: Harvard University Press.
47. Saylor Academy. (2012). The Coefficient of Determination. Retrieved from https://saylordotorg.github.io/text_introductory-statistics/s14-06-the-coefficient-of-determinati.html (Mar. 07, 2021)
48. Schroeder, M. R., & Gerlach, R. (1974). Diffusion, room shape and absorber location—influence on reverberation time. The Journal of the Acoustical Society of America, 56(4), 1300-1300.
49. Singh, A., Thakur, N., & Sharma, A. (2016, March). A review of supervised machine learning algorithms. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom) (pp. 1310-1315). Ieee.
50. Soni, J., Ansari, U., Sharma, D., & Soni, S. (2011). Predictive data mining for medical diagnosis: An overview of heart disease prediction. International Journal of Computer Applications, 17(8), 43-48.
51. Stephenson, U. (1985). Eine Schallteilchen-Computersimulation zur Berechnung der für die Hörsamkeit in Konzertsälen massgebenden Parameter. Acta Acustica united with Acustica, 59(1), 1-20.
52. Vorländer, M. (2008). Auralization: Fundamentals of Acoustics, Modelling, Simulation, Algorithms and Acoustic Virtual Reality. Springer International Publishing.
53. Wilson, A. C., Roelofs, R., Stern, M., Srebro, N., & Recht, B. (2017). The marginal value of adaptive gradient methods in machine learning. arXiv preprint arXiv:1705.08292.
54. Xiao, J., & Aletta, F. (2016). A soundscape approach to exploring design strategies for acoustic comfort in modern public libraries: A case study of the Library of Birmingham. Noise Mapping, 1(open-issue).
55. Yun, J. H., Ju, D. H., & Kim, J. S. (2007). Evaluation on Architectural Acoustic Performance of Small-scaled Multipurpose Hall for Improvement of Acoustic Performance. In Proceedings of the Korean Society for Noise and Vibration Engineering Conference (pp. 226-230). The Korean Society for Noise and Vibration Engineering.
56. Zhang, C., & Ma, Y. (Eds.). (2012). Ensemble machine learning: methods and applications. Springer Science & Business Media.
57. Zhang, Z. X. (2019). Boosting Algorithms Explained: Theory, Implementation, and Visualization. Retrieved from https://towardsdatascience.com/boosting-algorithms-explained-d38f56ef3f30 (Jan. 01, 2021).
58. Zhou, X., Späh, M., Hengst, K., & Zhang, T. (2021). Predicting the reverberation time in rectangular rooms with non-uniform absorption distribution. Applied Acoustics, 171, 107539.