Recurrent Neural Networks with Stochastic Layers for Acoustic Novelty Detection

Duong Nguyen (1, 2) , Oliver S. Kirsebom (3) , Fábio Frazão (3) , Ronan Fablet (2, 1) , Stan Matwin (3, 4)

1 IMT Atlantique - SC - Département Signal et Communications
2 Lab-STICC_IMTA_CID_TOMS
3 Institute for Big Data Analytics, Dalhousie
4 Institute of Computer Science, Polish Academy of Sciences

Mots clés: stochastic recurrent neural network; novelty detection; acoustic modeling; variational recurrent neural network

Domaines: Apprentissage [cs.LG]; Traitement du signal et de l'image [eess.SP]

Duong Nguyen, Oliver S. Kirsebom, Fábio Frazão, Ronan Fablet, Stan Matwin. Recurrent Neural Networks with Stochastic Layers for Acoustic Novelty Detection. ICASSP 2019 : International Conference on Acoustics, Speech, and Signal Processing, May 2019, Brighton, United Kingdom. ⟨10.1109/ICASSP.2019.8682901⟩. ⟨hal-02018277⟩