Investigating Label Noise Sensitivity of Convolutional Neural Networks for Fine Grained Audio Signal Labelling.
Rainer KelzGerhard WidmerPublished in: ICASSP (2018)
Keyphrases
- fine grained
- noise sensitivity
- audio signal
- convolutional neural networks
- coarse grained
- music information retrieval
- discrete wavelet transform
- access control
- pseudorandom
- speaker identification
- audio features
- compressed domain
- orientation estimation
- extracted features
- frequency domain
- wavelet transform
- multiresolution
- subband
- audio visual
- metadata
- text mining
- optical flow
- video sequences