Complete radiation balloons are needed to analyze the performance of loudspeakers and to perform accurate tunning and electroacoustic predictions of loudspeaker systems. A method is proposed to obtain full radiation balloons from horizontal and vertical polar measurements by using U-Net, a deep learning architecture widely applied for image processing. Mean absolute errors lower than 4 dB were obtained on test data.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=22548
(1363KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
