Recurrent Neural Networks with Stochastic Layers for Acoustic Novelty Detection

13 Feb 2019  ·  Duong Nguyen, Oliver S. Kirsebom, Fábio Frazão, Ronan Fablet, Stan Matwin ·

In this paper, we adapt Recurrent Neural Networks with Stochastic Layers, which are the state-of-the-art for generating text, music and speech, to the problem of acoustic novelty detection. By integrating uncertainty into the hidden states, this type of network is able to learn the distribution of complex sequences. Because the learned distribution can be calculated explicitly in terms of probability, we can evaluate how likely an observation is then detect low-probability events as novel. The model is robust, highly unsupervised, end-to-end and requires minimum preprocessing, feature engineering or hyperparameter tuning. An experiment on a benchmark dataset shows that our model outperforms the state-of-the-art acoustic novelty detectors.

PDF Abstract

Datasets


  Add Datasets introduced or used in this paper
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Acoustic Novelty Detection A3Lab PASCAL CHiME VRNN F1 93.6 # 2

Methods


No methods listed for this paper. Add relevant methods here