A novel approach for automatic acoustic novelty detection using a denoising autoencoder with bidirectional LSTM neural networks
Citations
310 citations
Cites background from "A novel approach for automatic acou..."
...DAEs, thus, provide a way to specify a noise model for ε (see Section II-C2), which has been applied for noise-robust acoustic novelty detection [42], for instance....
[...]
276 citations
257 citations
217 citations
Cites methods from "A novel approach for automatic acou..."
...In [15], novelty detection is performed for audio features using an auto-encoder with LSTM....
[...]
196 citations
Cites background from "A novel approach for automatic acou..."
...including acoustic signals [22], network server anomalies [33], data mining [14], document classification [21] and others....
[...]
References
7,290 citations
"A novel approach for automatic acou..." refers methods in this paper
...In addition to LSTM memory blocks, we use bidirectional RNNs [23]....
[...]
...Suitable types of networks for our purpose are RNNs and Bidirectional RNNs with LSTM units instead of ‘usual’ non-linear ones....
[...]
...The best network layout for our BRNNs has six hidden layers (three for each direction) with 216 LSTM units, each....
[...]
...The combination of bidirectional RNNs and LSTM memory blocks leads to bidirectional LSTM networks [24], where context from both temporal directions is exploited....
[...]
...The best network layout for our RNNs has three hidden layers with 156, 256, and 156 LSTM units, respectively....
[...]
5,303 citations
"A novel approach for automatic acou..." refers background in this paper
...The idea of denoising autoencoders [20] is quite intuitive....
[...]
4,814 citations
4,385 citations
"A novel approach for automatic acou..." refers background in this paper
...Deep neural networks use it during training of hidden layers to find common data representation from the input [18, 19]....
[...]
3,028 citations