Abstract
The purpose of this study is to predict the objective assessment of building acoustics more accurately and efficiently. In this paper, the neural network technology based on machine learning and computer acoustic simulation technology are combined to extract 10 typical characteristic parameters and 3 target parameters of 800 halls and rooms. Three matrix training sample databases are established by using Odeon platform. The reverberation time and speech transmission index are trained by BP neural network data fitting. The R results of the target parameters in this study are all more than 0.95. The MSE of the reverberation time parameter is in the range of 0.01-0.05 and the MSE of the STI parameter is less than 1 × 10−4. The results show that the neural network has good prediction accuracy, data generalization and application applicability. This prediction method can quickly evaluate the target parameters, reduce manpower and material resources, and improve work efficiency.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.