论文信息 - A Multi-modal Deep Neural Network approach to Bird-song Identication

A Multi-modal Deep Neural Network approach to Bird-song Identication

We present a multi-modal Deep Neural Network (DNN) approach for bird song identification. The presented approach takes both audio samples and metadata as input. The audio is fed into a Convolutional Neural Network (CNN) using four convolutional layers. The additionally provided metadata is processed using fully connected layers. The flattened convolutional layers and the fully connected layer of the metadata are joined and fed into a fully connected layer. The resulting architecture achieved 2., 3. and 4. rank in the BirdCLEF2017 task in various training configurations.

Thomas Lidy | Alexander Schindler | Botond Fazekas

[1] Thomas Hofmann,et al. Audio Based Bird Species Identification using Deep Learning Techniques , 2016, CLEF.

[2] G. Ball,et al. Neuroendocrine Mechanisms Regulating Reproductive Cycles and Reproductive Behavior in Birds , 2002 .

[3] T. C. van Flandern,et al. Low-precision formulae for planetary positions , 1979 .

[4] Thomas Lidy,et al. CQT-based Convolutional Neural Networks for Audio Scene Classification , 2016, DCASE.

[5] Andreas Rauber,et al. Comparing Shallow versus Deep Neural Network Architectures for Automatic Music Genre Classification , 2016, FMT.

[6] Hervé Glotin,et al. LifeCLEF 2016: Multimedia Life Species Identification Challenges , 2016, CLEF.

[7] Gregory F Ball,et al. Seasonal neuroplasticity in the songbird telencephalon: a role for melatonin. , 1999, Proceedings of the National Academy of Sciences of the United States of America.