CONF
He_ICASSP_2019/IDIAP
Adaptation of Multiple Sound Source Localization Neural Networks with Weak Supervision and Domain-Adversarial Training
He, Weipeng
Motlicek, Petr
Odobez, Jean-Marc
direction-of-arrival estimation
DOA estimation
domain adaptation
Encoding
Feature extraction
neural networks
Position measurement
Robots
sound source localization
training
weakly-supervised learning.
EXTERNAL
https://publications.idiap.ch/attachments/papers/2019/He_ICASSP_2019.pdf
PUBLIC
https://publications.idiap.ch/index.php/publications/showcite/He_Idiap-Com-01-2019
Related documents
Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
Brighton, United Kingdom
2019
770-774
2379-190X
978-1-4799-8131-1
10.1109/ICASSP.2019.8682655
doi
Despite the recent success of deep neural network-based approaches in sound source localization, these approaches suffer the limitations that the required annotation process is costly, and the mismatch between the training and test conditions undermines the performance. This paper addresses the question of how models trained with simulation can be exploited for multiple sound source localization in real scenarios by domain adaptation. In particular, two domain adaptation methods are investigated: weak supervision and domain-adversarial training. Our experiments show that the weak supervision with the knowledge of the number of sources can significantly improve the performance of an unadapted model. However, the domain-adversarial training does not yield significant improvement for this particular problem.
REPORT
He_Idiap-Com-01-2019/IDIAP
Adaptation of Multiple Sound Source Localization Neural Networks with Weak Supervision and Domain-Adversarial Training
He, Weipeng
Motlicek, Petr
Odobez, Jean-Marc
EXTERNAL
https://publications.idiap.ch/attachments/reports/2018/He_Idiap-Com-01-2019.pdf
PUBLIC
Idiap-Com-01-2019
2019
Idiap
June 2019