Residual Recurrent Neural Network for Speech Enhancement

Abdulbaqi, Jalal; Gu, Yue; Chen, Shuhong; Marsic, Ivan

doi:10.1109/icassp40776.2020.9053544

Cited by 16 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The STFT loss is introduced in [ 34 ]. There are examples that use log-cosh as the loss function for the speech enhancement model, one built with GRU [ 41 ] and the other built with SRU [ 20 ]. We creatively used one of the evaluation metrics, specifically SNR, as a term in the loss function.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Waveform Mapping-Based Approach for Enhancement of Trunk Borers’ Vibration Signals Using Deep Learning Model

Shi

Chen

Zhang

et al. 2022

Insects

View full text Add to dashboard Cite

The larvae of some trunk-boring beetles barely leave traces on the outside of trunks when feeding within, rendering the detection of them rather difficult. One approach to solving this problem involves the use of a probe to pick up boring vibrations inside the trunk and distinguish larvae activity according to the vibrations. Clean boring vibration signals without noise are critical for accurate judgement. Unfortunately, these environments are filled with natural or artificial noise. To address this issue, we constructed a boring vibration enhancement model named VibDenoiser, which makes a significant contribution to this rarely studied domain. This model is built using the technology of deep learning-based speech enhancement. It consists of convolutional encoder and decoder layers with skip connections, and two layers of SRU++ for sequence modeling. The dataset constructed for study is made up of boring vibrations of Agrilus planipennis Fairmaire, 1888 (Coleoptera: Buprestidae) and environmental noise. Our VibDenoiser achieves an improvement of 18.57 in SNR, and it runs in real-time on a laptop CPU. The accuracy of the four classification models increased by a large margin using vibration clips enhanced by our model. The results demonstrate the great enhancement performance of our model, and the contribution of our work to better boring vibration detection.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The STFT loss is introduced in [34]. There are examples that use log-cosh as the loss function for the speech enhancement model, one built with GRU [41] and the other built with SRU [20].…”

Section: Loss Functionmentioning

confidence: 99%

A Waveform Mapping-Based Approach for Enhancement of Trunk Borers’ Vibration Signals Using Deep Learning Model

Shi

Chen

Zhang

et al. 2022

Insects

View full text Add to dashboard Cite

show abstract

“…The study [ 31 ] presents a completely E2E recurrent neural network (RNN) for enhancing single-channel speech. By lowering the feature resolution without sacrificing the information, an hourglass-shaped network effectively captured long-range temporal correlation.…”

Section: Related Studiesmentioning

confidence: 99%

End-to-End Deep Convolutional Recurrent Models for Noise Robust Waveform Speech Enhancement

Ullah

Wuttisittikulkij

Chaudhary

et al. 2022

Sensors

View full text Add to dashboard Cite

Because of their simple design structure, end-to-end deep learning (E2E-DL) models have gained a lot of attention for speech enhancement. A number of DL models have achieved excellent results in eliminating the background noise and enhancing the quality as well as the intelligibility of noisy speech. Designing resource-efficient and compact models during real-time processing is still a key challenge. In order to enhance the accomplishment of E2E models, the sequential and local characteristics of speech signal should be efficiently taken into consideration while modeling. In this paper, we present resource-efficient and compact neural models for end-to-end noise-robust waveform-based speech enhancement. Combining the Convolutional Encode-Decoder (CED) and Recurrent Neural Networks (RNNs) in the Convolutional Recurrent Network (CRN) framework, we have aimed at different speech enhancement systems. Different noise types and speakers are used to train and test the proposed models. With LibriSpeech and the DEMAND dataset, the experiments show that the proposed models lead to improved quality and intelligibility with fewer trainable parameters, notably reduced model complexity, and inference time than existing recurrent and convolutional models. The quality and intelligibility are improved by 31.61% and 17.18% over the noisy speech. We further performed cross corpus analysis to demonstrate the generalization of the proposed E2E SE models across different speech datasets.

show abstract

“…The feed-forward network estimates the compatibility function with a single hidden layer (Vaswani et al 2017). Different from the conventional Transformer encoder, in the feed-forward network, the first fully connected layer is replaced by a gated recurrent unit (GRU) (Abdulbaqi et al 2020) layer because the GRU has a simpler structure and trains faster than local LSTMs. Moreover, the same dimension of attentions maps is obtained from the input and output of one sub-layer in the U-Transformer, e.g., d layer = 512 in the first sub-layer, to facilitate the residual connections.…”

Section: Speech Enhancement U-transformermentioning

confidence: 99%

U-shaped Transformer with Frequency-Band Aware Attention for Speech Enhancement

Li¹,

Sun²,

Naqvi³

2021

Preprint

View full text Add to dashboard Cite

The state-of-the-art speech enhancement has limited performance in speech estimation accuracy. Recently, in deep learning, the Transformer shows the potential to exploit the longrange dependency in speech by self-attention. Therefore, it is introduced in speech enhancement to improve the speech estimation accuracy from a noise mixture. However, to address the computational cost issue in Transformer with selfattention, the axial attention is the option i.e., to split a 2D attention into two 1D attentions. Inspired by the axial attention, in the proposed method we calculate the attention map along both time-and frequency-axis to generate time and frequency sub-attention maps. Moreover, different from the axial attention, the proposed method provides two parallel multi-head attentions for time-and frequency-axis. Furthermore, it is proven in the literature that the lower frequencyband in speech, generally, contains more desired information than the higher frequency-band, in a noise mixture. Therefore, the frequency-band aware attention is proposed i.e., high frequency-band attention (HFA), and low frequency-band attention (LFA). The U-shaped Transformer is also first time introduced in the proposed method to further improve the speech estimation accuracy. The extensive evaluations over four public datasets, confirm the efficacy of the proposed method.

show abstract

Residual Recurrent Neural Network for Speech Enhancement

Cited by 16 publications

References 31 publications

A Waveform Mapping-Based Approach for Enhancement of Trunk Borers’ Vibration Signals Using Deep Learning Model

A Waveform Mapping-Based Approach for Enhancement of Trunk Borers’ Vibration Signals Using Deep Learning Model

End-to-End Deep Convolutional Recurrent Models for Noise Robust Waveform Speech Enhancement

U-shaped Transformer with Frequency-Band Aware Attention for Speech Enhancement

Contact Info

Product

Resources

About