Machine-learning-based estimation of reverberation time using room geometry for room effect rendering

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Researchers

Research units

Abstract

This work presents a machine-learning-based method to estimate the reverberation time of a virtual room for auralization purposes. The models take as input geometric features of the room and output the estimated reverberation time values as function of frequency. The proposed model is trained and evaluated using a novel dataset composed of real-world acoustical measurements of a single room with 832 different configurations of furniture and absorptive materials, for multiple loudspeaker positions. The method achieves a prediction accuracy
of approximately 90% for most frequency bands. Furthermore, when comparing against the Sabine and Eyring methods, the proposed approach exhibits a much higher accuracy, especially at low frequencies.

Details

Original languageEnglish
Title of host publicationProceedings of the 23rd International Congress on Acoustics : integrating 4th EAA Euroregio 2019 : 9-13 September 2019 in Aachen, Germany
Publication statusPublished - 13 Sep 2019
MoE publication typeA4 Article in a conference publication
EventInternational Congress on Acoustics - Aachen, Germany
Duration: 9 Sep 201913 Sep 2019
Conference number: 23

Conference

ConferenceInternational Congress on Acoustics
Abbreviated titleICA
CountryGermany
CityAachen
Period09/09/201913/09/2019

    Research areas

  • reverberation time, machine learning, room acoustics

ID: 40315145