Automatic audio source classification system for recordings captured with microphone array

Chrul M, Ruminski A, Zernicki T, Lukasik E. Automatic audio source classification system for recordings captured with microphone array. In: AES Conference: 2020 AES International Conference on Audio for Virtual and Augmented Reality (August 2020). Audio Engineering Society; 2020. Paper 10481. Available from: https://aes.org/publications/elibrary-page/?id=21158

@inproceedings{Chrul2020_21158,
  author = {Chrul, Michal and Ruminski, Andrzej and Zernicki, Tomasz and Lukasik, Ewa},
  title = {{Automatic audio source classification system for recordings captured with microphone array}},
  booktitle = {AES Conference: 2020 AES International Conference on Audio for Virtual and Augmented Reality (August 2020)},
  note = {Paper 10481},
  year = {2020},
  month = aug,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=21158}
}

TY  - CPAPER
TI  - Automatic audio source classification system for recordings captured with microphone array
AU  - Chrul, Michal
AU  - Ruminski, Andrzej
AU  - Zernicki, Tomasz
AU  - Lukasik, Ewa
T2  - AES Conference: 2020 AES International Conference on Audio for Virtual and Augmented Reality (August 2020)
M1  - Paper 10481
PY  - 2020
DA  - 2020/08/06
UR  - https://aes.org/publications/elibrary-page/?id=21158
PB  - Audio Engineering Society
LA  - en
AB  - The aim of this paper was to create an automatic sound source classification framework for recordings captured with a microphone array and evaluate the sound source separation algorithm impact on the classification results. The preprocessing related to the said evaluation concerned convolving the dataset samples with impulse responses captured with a microphone array, as well as mixing the samples together to simulate their co-presence in a virtual recording scene. This way, the evaluation of the separation algorithm impact on classification results was possible. Furthermore, such approach saved multiple hours of labour that would need to be spent on the recording process itself. Finally, the classification results delivered by different models were evaluated and compared.
ER  -