%Aigaion2 BibTeX export from Idiap Publications
%Friday 05 December 2025 03:34:06 AM

@INPROCEEDINGS{He_ICRA_2018,
                      author = {He, Weipeng and Motlicek, Petr and Odobez, Jean-Marc},
                    keywords = {acoustic generators, Artificial Neural Networks, deep neural networks, Delays, Encoding, Estimation, human-robot interaction, likelihood-based encoding, microphone arrays, Microphones, multiple sound sources, multiple speaker detection, network output, neural nets, neural network-based sound source localization methods, Robots, simultaneous detection, single sound source, sound mixtures, spatial spectrum-based approaches, speaker recognition},
                    projects = {Idiap, MUMMER},
                       month = may,
                       title = {Deep Neural Networks for Multiple Speaker Detection and Localization},
                   booktitle = {2018 IEEE International Conference on Robotics and Automation (ICRA)},
                        year = {2018},
                       pages = {74-79},
                    location = {Brisbane, AUSTRALIA},
                        issn = {1050-4729},
                        isbn = {978-1-5386-3081-5},
                         doi = {10.1109/ICRA.2018.8461267},
                    crossref = {He_Idiap-RR-02-2018},
                    abstract = {We propose to use neural networks for simultaneous detection and localization of multiple sound sources in human-robot interaction. In contrast to conventional signal processing techniques, neural network-based sound source localization methods require fewer strong assumptions about the environment. Previous neural network-based methods have been focusing on localizing a single sound source, which do not extend to multiple sources in terms of detection and localization. In this paper, we thus propose a likelihood-based encoding of the network output, which naturally allows the detection of an arbitrary number of sources. In addition, we investigate the use of sub-band cross-correlation information as features for better localization in sound mixtures, as well as three different network architectures based on different motivations. Experiments on real data recorded from a robot show that our proposed methods significantly outperform the popular spatial spectrum-based approaches.},
                         pdf = {https://publications.idiap.ch/attachments/papers/2019/He_ICRA_2018.pdf}
}



crossreferenced publications: 
@TECHREPORT{He_Idiap-RR-02-2018,
                      author = {He, Weipeng and Motlicek, Petr and Odobez, Jean-Marc},
                    projects = {Idiap, MUMMER},
                       month = {2},
                       title = {Deep Neural Networks for Multiple Speaker Detection and Localization},
                        type = {Idiap-RR},
                      number = {Idiap-RR-02-2018},
                        year = {2018},
                 institution = {Idiap},
                         pdf = {https://publications.idiap.ch/attachments/reports/2017/He_Idiap-RR-02-2018.pdf}
}