Comparison of Performance in Binaural Sound Source Localisation using Convolutional Neural Networks for differing Feature Representations

Reed-Jones JT, Jones KO, Fergus P, Marsland J, Ellis DL. Comparison of Performance in Binaural Sound Source Localisation using Convolutional Neural Networks for differing Feature Representations. In: AES Convention 154. Audio Engineering Society; 2023.pp. 59:1-. Available from: https://aes.org/publications/elibrary-page/?id=22084

@inproceedings{ReedJones2023_22084,
  author = {Reed-Jones, Jago T. and Jones, Karl O. and Fergus, Paul and Marsland, John and Ellis, David L.},
  title = {{Comparison of Performance in Binaural Sound Source Localisation using Convolutional Neural Networks for differing Feature Representations}},
  booktitle = {AES Convention 154},
  pages = {59:1-},
  year = {2023},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=22084}
}

TY  - CPAPER
TI  - Comparison of Performance in Binaural Sound Source Localisation using Convolutional Neural Networks for differing Feature Representations
AU  - Reed-Jones, Jago T.
AU  - Jones, Karl O.
AU  - Fergus, Paul
AU  - Marsland, John
AU  - Ellis, David L.
T2  - AES Convention 154
SP  - 1
M1  - Paper 59
PY  - 2023
DA  - 2023/05/06
UR  - https://aes.org/publications/elibrary-page/?id=22084
PB  - Audio Engineering Society
LA  - en
AB  - Binaural Sound Source Localisation is increasingly being achieved by means of the Convolutional Neural Network (CNN). These networks take in a Time-Frequency representation of audio as an input, and use this to estimate the direction of arrival of a sound. In previous works, different Time-Frequency representations have been used, but never only using solely magnitude spectra, leading to a lack of understanding in the importance of this in full azimuthal binaural sound source localisation. This work aims to address that gap by testing the performance of a CNN trained and tested on four different Time-Frequency representations: Mel-Spectrogram, Gammatonegram, Mel-Frequency Cepstrum, and Gammatone-Frequency Cepstrum. From this test, it was found that Spectrograms are suitable for the task of full azimuthal binaural sound source localisation.
ER  -