The sound of thunder is widely used in game, film and virtual reality sound design. It is also a phenomenon for which we seek a better understanding of the physics underlying the sound. Though many models of thunder have been proposed, there has not yet been a formal perceptual evaluation of the models to assess their realism and sound quality. Here, we present and evaluate the implementation of several thunder sound effect synthesis models. The models include different physical modeling and signal-based approaches, as well as a recorded sample. Evaluation was with over 50 participants. The results showed that none of the models were close to the recording in terms of realism, though signal-based models slightly outperformed the physical models. This highlights the need for comparative perceptual evaluation in sound synthesis, and identifies the limitations of current thunder simulation approaches.
(See document for exact affiliation information.)
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.
