Spectral Phase Estimation Based on Deep Neural Networks for Single Channel Speech Enhancement

Saleem, Nasir; Khattak, Muhammad Irfan; Perez, Eliud

doi:10.1134/s1064226919120155

Cited by 8 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper emphasize the speech magnitude enhancement. We will be devoted to include the phase estimation [ 57 ] and incorporate with the proposed SE model. Moreover, more robust loss functions will be worked out for better results.…”

Section: Discussionmentioning

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

Section: Discussionmentioning

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

show abstract

“…In the AEC system, the adaptive filter is employed to identify the room IR. The filter needs to be updated continuously, because the characteristics of the room vary in time with the movement of people and objects [4,5]. In Figure 1, 𝑠 𝑏 (𝑛) is convoluted by IR ℎ 𝑎 (𝑛) of room, and then captured, simultaneously, by a microphone with the 𝑠 𝑎 (𝑛).…”

Section: Acoustic Echo Cancellation Systemmentioning

confidence: 99%

“…Recently, several research papers have been published based on adaptive filtering algorithms, which have been implemented on several telecommunication fields, such as acoustic echo and noise reduction to accelerate the convergence adaptation and enhancing the quality of conversations [1][2][3]. Adaptive filtering algorithms are largely used in acoustic echo cancellation [4][5][6]. Several researches have been conducted to address the identification problem in the time domain and in the frequency domain [7,8].…”

Section: Introductionmentioning

confidence: 99%

New Variable Selected Coefficients Adaptive Sparse Algorithm for Acoustic System Identification

Hassani,

Bendoumia,

Guessoum

et al. 2024

View full text Add to dashboard Cite

“…The findings revealed that gammatone features is one of the outperform features. Various speech dataset such as IEEE [10,14,15,17] and TIMIT [18] were used to analyze the performance, which normally in short duration of speech utterance. A study in [10] also compared the training targets in DNN algorithm.…”

Section: Related Workmentioning

confidence: 99%

“…Some studies proposed a new target mask in the DNN such as less aggressive Wiener filtering, phase aware, and complex ratio mask [20][21][22]. Other studies also proposed an alternative approach using postfiltering techniques such as global variance equalization [17,24] after the DNN-based mask estimation and speech reconstruction.…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Approach for Single Channel Speech Enhancement using Deep Neural Network and Harmonic Regeneration Noise Reduction

Jamal¹,

Fuad²,

Sha’abani³

2020

IJACSA

View full text Add to dashboard Cite

This paper presents a hybrid approach for single channel speech enhancement using deep neural network (DNN) and harmonic regeneration noise reduction (HRNR). The DNN was used as a supervised algorithm to predict new target mask such as constrained Wiener Filter (cWF) target mask from noisy mixture signal that was transformed into gammatone filter bank features. Meanwhile, HRNR algorithm was applied in the postfiltering strategy to eliminate residual noise. The DNN algorithm is an emerging supervised speech enhancement to overcome heavy nonstationary noise and low signal-to-noise ratio (SNR) issues. To validate the proposed algorithm with new target mask, 600 Malay utterances combining male and female speakers were used in a training session while 120 Malay utterances were used in a prediction session. The short time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ) scores were calculated as the performance metrics. In this work, the proposed target mask outperformed other baseline target masks. Thus, PESQ and STOI scores for the hybrid speech enhancement algorithm is 1.17 and 0.79, respectively, at -5 dB babble noise SNR.

show abstract

Spectral Phase Estimation Based on Deep Neural Networks for Single Channel Speech Enhancement

Cited by 8 publications

References 26 publications

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

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

New Variable Selected Coefficients Adaptive Sparse Algorithm for Acoustic System Identification

A Hybrid Approach for Single Channel Speech Enhancement using Deep Neural Network and Harmonic Regeneration Noise Reduction

Contact Info

Product

Resources

About