Binaural technology becomes increasingly popular in the multimedia systems. This paper identifies a set of features of binaural recordings suitable for the automatic classification of the four basic spatial audio scenes representing the most typical patterns of audio content distribution around a listener. Moreover, it compares the five artificial-intelligence-based methods applied to the classification of binaural recordings. The results show that both the spatial and the spectro-temporal features are essential to accurate classification of binaurally rendered acoustic scenes. The spectro-temporal features appear to have a stronger influence on the classification results than the spatial metrics. According to the obtained results, the method based on the support vector machine, exploiting the features identified in the study, yields the classification accuracy approaching 84%.
|Title of host publication||IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018|
|Editors||Maria Ganzha, Leszek Maciaszek, Marcin Paprzycki|
|Number of pages||4|
|Publication status||Published - 29 Oct 2018|
|Event||IEEE Federated Conference on Computer Science and Information Systems|
- Poznan, Poland
Duration: 9 Sep 2018 → 12 Sep 2018
https://fedcsis.org/2018/ (Link to Conference Website)
|Conference||IEEE Federated Conference on Computer Science and Information Systems|
|Period||9/09/18 → 12/09/18|
Zieliński, S., & Lee, H. (2018). Feature Extraction of Binaural Recordings for Acoustic Scene Classification. In M. Ganzha, L. Maciaszek, & M. Paprzycki (Eds.), IEEE Federated Conference on Computer Science and Information Systems, FedCSIS 2018 (pp. 585-588). IEEE.