scholarly journals Polyphonic Sound Event Detection Based on Residual Convolutional Recurrent Neural Network with Semi-supervised Loss Function

IEEE Access ◽  
2021 ◽  
pp. 1-1
Author(s):  
Nam Kyun Kim ◽  
Hong Kook Kim
Keyword(s):  
Neural Network ◽  
Loss Function ◽  
Recurrent Neural Network ◽  
Event Detection ◽  
Sound Event ◽  
Sound Event Detection

scholarly journals Forward-Backward Convolutional Recurrent Neural Networks and Tag-Conditioned Convolutional Neural Networks for \\Weakly Labeled Semi-supervised Sound Event Detection

2021 ◽  
Author(s):  
Janek Ebbers ◽  
Reinhold Haeb-Umbach
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Recurrent Neural Network ◽  
Event Detection ◽  
Forward Direction ◽  
Time Step ◽  
Sound Event ◽  
Sound Event Detection ◽  
Validation Set ◽  
Classi Fication

In this paper we present our system for thedetection and classi-fication of acoustic scenes and events (DCASE) 2020 ChallengeTask 4: Sound event detection and separation in domestic envi-ronments. We introduce two new models: the forward-backwardconvolutional recurrent neural network (FBCRNN) and the tag-conditioned convolutional neural network (CNN). The FBCRNNemploys two recurrent neural network (RNN) classifiers sharing thesame CNN for preprocessing. With one RNN processing a record-ing in forward direction and the other in backward direction, thetwo networks are trained to jointly predict audio tags, i.e., weak la-bels, at each time step within a recording, given that at each timestep they have jointly processed the whole recording. The pro-posed training encourages the classifiers to tag events as soon aspossible. Therefore, after training, the networks can be appliedto shorter audio segments of, e.g.,200 ms, allowing sound eventdetection (SED). Further, we propose a tag-conditioned CNN tocomplement SED. It is trained to predict strong labels while using(predicted) tags, i.e., weak labels, as additional input. For train-ing pseudo strong labels from a FBCRNN ensemble are used. Thepresented system scored the fourth and third place in the systemsand teams rankings, respectively. Subsequent improvements allowour system to even outperform the challenge baseline and winnersystems in average by, respectively,18.0 %and2.2 %event-basedF1-score on the validation set. Source code is publicly available athttps://github.com/fgnt/pb_sed

scholarly journals MTF-CRNN: Multiscale Time-Frequency Convolutional Recurrent Neural Network for Sound Event Detection

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 147337-147348
Author(s):  
Keming Zhang ◽  
Yuanwen Cai ◽  
Yuan Ren ◽  
Ruida Ye ◽  
Liang He
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Event Detection ◽  
Time Frequency ◽  
Sound Event ◽  
Sound Event Detection

scholarly journals Joint Optimization of Deep Neural Network-Based Dereverberation and Beamforming for Sound Event Detection in Multi-Channel Environments

Sensors ◽  
2020 ◽  
Vol 20 (7) ◽  
pp. 1883 ◽  
Author(s):  
Kyoungjin Noh ◽  
Joon-Hyuk Chang
Keyword(s):  
Neural Network ◽  
Loss Function ◽  
Event Detection ◽  
Deep Neural Network ◽  
Joint Optimization ◽  
Audio Signals ◽  
Sound Event ◽  
Minimum Variance Distortionless Response ◽  
Reverberant Environments ◽  
Sound Event Detection

In this paper, we propose joint optimization of deep neural network (DNN)-supported dereverberation and beamforming for the convolutional recurrent neural network (CRNN)-based sound event detection (SED) in multi-channel environments. First, the short-time Fourier transform (STFT) coefficients are calculated from multi-channel audio signals under the noisy and reverberant environments, which are then enhanced by the DNN-supported weighted prediction error (WPE) dereverberation with the estimated masks. Next, the STFT coefficients of the dereverberated multi-channel audio signals are conveyed to the DNN-supported minimum variance distortionless response (MVDR) beamformer in which DNN-supported MVDR beamforming is carried out with the source and noise masks estimated by the DNN. As a result, the single-channel enhanced STFT coefficients are shown at the output and tossed to the CRNN-based SED system, and then, the three modules are jointly trained by the single loss function designed for SED. Furthermore, to ease the difficulty of training a deep learning model for SED caused by the imbalance in the amount of data for each class, the focal loss is used as a loss function. Experimental results show that joint training of DNN-supported dereverberation and beamforming with the SED model under the supervision of focal loss significantly improves the performance under the noisy and reverberant environments.

scholarly journals Multi-scale Convolutional Recurrent Neural Network and Data Augmentation for Polyphonic Sound Event Detection

2021 ◽  
Vol 1769 (1) ◽  
pp. 012008
Author(s):  
Keming Zhang ◽  
Yuanwen Cai ◽  
Yuan Ren ◽  
Ruida Ye ◽  
Xianwei Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Event Detection ◽  
Data Augmentation ◽  
Multi Scale ◽  
Sound Event ◽  
Sound Event Detection
Sign in / Sign up

Export Citation Format

Share Document