Nonlinear enhancement of noisy speech, using continuous attractor dynamics formed in recurrent neural networks

Authors:
Louiza Dehyadegary;Seyyed Ali Seyyedsalehi;Isar Nejadgholi
Affiliations:
Department of Biomedical Engineering, Amirkabir University of Technology, 424, Hafez Ave, Tehran, Iran;Department of Biomedical Engineering, Amirkabir University of Technology, 424, Hafez Ave, Tehran, Iran;Department of Biomedical Engineering, Amirkabir University of Technology, 424, Hafez Ave, Tehran, Iran
Venue:
Neurocomputing
Year:
2011

Citing 7
Cited 1

Speech recognition by machines and humans

Speech Communication
Robust automatic speech recognition with missing and unreliable acoustic data

Speech Communication
Toward Robust Speech Recognition and Understanding

Journal of VLSI Signal Processing Systems
Attractor Dynamics in Feedforward Neural Networks

Neural Computation
Non-linear PCA: a missing data approach

Bioinformatics
Nonlinear normalization of input patterns to speaker variability in speech recognition neural networks

Neural Computing and Applications
Coordinated training of noise removing networks

ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: plenary, special, audio, underwater acoustics, VLSI, neural networks - Volume I

Real-time frequency-based noise-robust Automatic Speech Recognition using Multi-Nets Artificial Neural Networks: A multi-views multi-learners approach

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

Here, formation of continuous attractor dynamics in a nonlinear recurrent neural network is used to achieve a nonlinear speech denoising method, in order to implement robust phoneme recognition and information retrieval. Formation of attractor dynamics in recurrent neural network is first carried out by training the clean speech subspace as the continuous attractor. Then, it is used to recognize noisy speech with both stationary and nonstationary noise. In this work, the efficiency of a nonlinear feedforward network is compared to the same one with a recurrent connection in its hidden layer. The structure and training of this recurrent connection, is designed in such a way that the network learns to denoise the signal step by step, using properties of attractors it has formed, along with phone recognition. Using these connections, the recognition accuracy is improved 21% for the stationary signal and 14% for the nonstationary one with 0db SNR, in respect to a reference model which is a feedforward neural network.