Learning to Dequantize Speech Signals by Primal-dual Networks: an Approach for Acoustic Sensor Networks

Brauer, Christoph; Zhao, Ziyue; Lorenz, Dirk A.; Fingscheidt, Tim

doi:10.1109/icassp.2019.8683341

Cited by 9 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…15. This loss has shown superior performance compared to the MSE loss in speech enhancement [29], as well as for quantized speech reconstruction [30]. Some more detail is given in Appendix 1.…”

Section: Baseline Pw-filtmentioning

confidence: 99%

“…The parameters of the deep learning architectures are then optimized by minimizing the MSE between the inferred results and their corresponding targets. In reality, optimization of the MSE loss in training does not guarantee any perceptual quality of the speech component and of the residual noise component, respectively, which leads to limited performance [27][28][29][30][31][32][33][34][35][36]. This effect is even more evident when the level of the noise component is significantly higher than that of the speech component in some regions of the noisy speech spectrum, which explains the bad performance at lower SNR conditions when training with MSE.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Components loss for neural networks in mask-based speech enhancement

Elshamy

Zhao

et al. 2021

J AUDIO SPEECH MUSIC PROC.

Self Cite

View full text Add to dashboard Cite

Estimating time-frequency domain masks for single-channel speech enhancement using deep learning methods has recently become a popular research field with promising results. In this paper, we propose a novel components loss (CL) for the training of neural networks for mask-based speech enhancement. During the training process, the proposed CL offers separate control over preservation of the speech component quality, suppression of the noise component, and preservation of a naturally sounding residual noise component. We illustrate the potential of the proposed CL by evaluating a standard convolutional neural network (CNN) for mask-based speech enhancement. The new CL is compared to several baseline losses, comprising the conventional mean squared error (MSE) loss w.r.t. speech spectral amplitudes or w.r.t. an ideal-ratio mask, auditory-related loss functions, such as the perceptual evaluation of speech quality (PESQ) loss and the perceptual weighting filter loss, and also the recently proposed SNR loss with two masks. Detailed analysis suggests that the proposed CL obtains a better or at least a more balanced performance across all employed instrumental quality metrics, including SNR improvement, speech component quality, enhanced total speech quality, and particularly also delivers a natural sounding residual noise component. For unseen noise types, we excel even perceptually motivated losses by an about 0.2 points higher PESQ score. The recently proposed so-called SNR loss with two masks not only requires a network with more parameters due to the two decoder heads, but also falls behind on PESQ and POLQA and particularly w.r.t. residual noise quality. Note that the proposed CL shows significantly more 1st ranks among the evaluation metrics than any other baseline. It is easy to implement, and code is provided at https://github.com/ifnspaml/Components-Loss.

show abstract

Section: Baseline Pw-filtmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Components loss for neural networks in mask-based speech enhancement

Elshamy

Zhao

et al. 2021

J AUDIO SPEECH MUSIC PROC.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The perceptual weighting filter is originally used to shape the coding noise / quantization error to be less audible by exploiting the masking property of the human ear [27]. Motivated by the success of the weighting filter when used to shape coding noise, we apply it in this work to design the loss function to achieve improved perceptual quality in the context of speech enhancement, aiming to combat acoustic background noise (instead of coding noise, as has been done in [28]). We propose to extract the frequency response of the weighting filter from the clean speech and directly apply it to the loss function, resulting in an unaltered DNN architecture with no need to change the topology as in [23].…”

mentioning

confidence: 99%

A Perceptual Weighting Filter Loss for DNN Training In Speech Enhancement

Zhao

Elshamy

Fingscheidt

2019

2019 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)

Self Cite

View full text Add to dashboard Cite

Single-channel speech enhancement with deep neural networks (DNNs) has shown promising performance and is thus intensively being studied. In this paper, instead of applying the mean squared error (MSE) as the loss function during DNN training for speech enhancement, we design a perceptual weighting filter loss motivated by the weighting filter as it is employed in analysis-by-synthesis speech coding, e.g., in code-excited linear prediction (CELP). The experimental results show that the proposed simple loss function improves the speech enhancement performance compared to a reference DNN with MSE loss in terms of perceptual quality and noise attenuation. The proposed loss function can be advantageously applied to an existing DNN-based speech enhancement system, without modification of the DNN topology for speech enhancement.

show abstract

“…Even more recently, a deep-learning-based method has been published [16]. Nonetheless, the classical approaches have not yet been fully explored, which is why the goal of this paper is to evaluate a number of sparsity-based audio dequantization methods.…”

Section: Introductionmentioning

confidence: 99%

Audio Dequantization Using (Co)Sparse (Non)Convex Methods

Záviška

Rajmic

Mokrý

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

The paper deals with the hitherto neglected topic of audio dequantization. It reviews the state-of-the-art sparsity-based approaches and proposes several new methods. Convex as well as non-convex approaches are included, and all the presented formulations come in both the synthesis and analysis variants.In the experiments the methods are evaluated using the signalto-distortion ratio (SDR) and PEMO-Q, a perceptually motivated metric.

show abstract

Learning to Dequantize Speech Signals by Primal-dual Networks: an Approach for Acoustic Sensor Networks

Cited by 9 publications

References 15 publications

Components loss for neural networks in mask-based speech enhancement

Components loss for neural networks in mask-based speech enhancement

A Perceptual Weighting Filter Loss for DNN Training In Speech Enhancement

Audio Dequantization Using (Co)Sparse (Non)Convex Methods

Contact Info

Product

Resources

About