Icassp 2022 Deep Noise Suppression Challenge

Dubey, Harishchandra; Gopal, Vishak; Cutler, Ross; Aazami, Ashkan; Matusevych, Sergiy; Braun, Sebastian; Eskimez, Şefik Emre; Thakker, Manthan; Yoshioka, Takuya; Gamper, Hannes; Aichner, Robert

doi:10.1109/icassp43922.2022.9747230

Cited by 95 publications

(52 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed FRCRN model achieves state-of-the-art (SOTA) results on the DNS-2020 dataset [11] and the Voicebank+Demand dataset [12]. Our submission to the ICASSP 2022 Deep Noise Suppression (DNS) challenge (DNS-2022) [13] ranked overall 2nd place for the real-time fullband nono-personalized track in terms of Mean Opinion Score (MOS) and Word Accuracy (WAcc).…”

Section: Introductionmentioning

confidence: 86%

See 1 more Smart Citation

FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement

Zhao

Watcharasupat

et al. 2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Convolutional recurrent networks (CRN) integrating a convolutional encoder-decoder (CED) structure and a recurrent structure have achieved promising performance for monaural speech enhancement. However, feature representation across frequency context is highly constrained due to limited receptive fields in the convolutions of CED. In this paper, we propose a convolutional recurrent encoderdecoder (CRED) structure to boost feature representation along the frequency axis. The CRED applies frequency recurrence on 3D convolutional feature maps along the frequency axis following each convolution, therefore, it is capable of catching long-range frequency correlations and enhancing feature representations of speech inputs. The proposed frequency recurrence is realized efficiently using a feedforward sequential memory network (FSMN). Besides the CRED, we insert two stacked FSMN layers between the encoder and the decoder to model further temporal dynamics. We name the proposed framework as Frequency Recurrent CRN (FRCRN). We design FRCRN to predict complex Ideal Ratio Mask (cIRM) in complex-valued domain and optimize FRCRN using both time-frequency-domain and time-domain losses. Our proposed approach achieved state-of-the-art performance on wideband benchmark datasets and achieved 2nd place for the real-time fullband track in terms of Mean Opinion Score (MOS) and Word Accuracy (WAcc) in the ICASSP 2022 Deep Noise Suppression (DNS) challenge.

show abstract

Section: Introductionmentioning

confidence: 86%

“…We further extended the evaluation of our proposed FRCRN model on the fullband DNS-2022 dataset [13]. For the training setup, we keep the same window length of 20ms and frame shift of 10ms.…”

Section: Evaluation On Fullband Speech Signalmentioning

confidence: 99%

FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement

Zhao

Watcharasupat

et al. 2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…However, the background noise and reverberation degrade the speech and transcription quality. As these acoustic distortions can hamper communications and thus productivity, the speech enhancement (SE) field has drawn a lot of renewed attention recently [1,2].…”

Section: Introductionmentioning

confidence: 99%

“…Personalized speech enhancement (PSE) provides an improvement to the general SE approach by using prior knowledge about a target speaker [2,3,4,5]. One exemplary approach to PSE is to extract a speaker embedding vector from a short enrollment audio sample of the target speaker and feed it to an SE model.…”

Section: Introductionmentioning

confidence: 99%

Fast Real-time Personalized Speech Enhancement: End-to-End Enhancement Network (E3Net) and Knowledge Distillation

Thakker¹,

Eskimez²,

Yoshioka³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This paper investigates how to improve the runtime speed of personalized speech enhancement (PSE) networks while maintaining the model quality. Our approach includes two aspects: architecture and knowledge distillation (KD). We propose an end-to-end enhancement (E3Net) model architecture, which is 3× faster than a baseline STFT-based model. Besides, we use KD techniques to develop compressed student models without significantly degrading quality. In addition, we investigate using noisy data without reference clean signals for training the student models, where we combine KD with multi-task learning (MTL) using an automatic speech recognition (ASR) loss. Our results show that E3Net provides better speech and transcription quality with a lower target speaker over-suppression (TSOS) rate than the baseline model. Furthermore, we show that the KD methods can yield student models that are 2 − 4× faster than the teacher and provides reasonable quality. Combining KD and MTL improves the ASR and TSOS metrics without degrading the speech quality.

show abstract

“…This is evident by the increasingly large amount of research continuously attempting to push the performance boundaries of current SE systems [5], [6]. The majority of these approaches harness the recent advances in deep learning (DL) techniques as well as the increasingly more available speech datasets [7]- [10].…”

Section: Introductionmentioning

confidence: 99%

CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement

Abdulatif¹,

Cao²,

Yang³

2022

Preprint

View full text Add to dashboard Cite

<p>Convolution-augmented transformers (Conformers) are recently proposed in various speech-domain applications, such as automatic speech recognition (ASR) and speech separation, as they can capture both local and global dependencies. In this paper, we propose a conformer-based metric generative adversarial network (CMGAN) for speech enhancement (SE) in the time-frequency (TF) domain. The generator encodes the magnitude and complex spectrogram information using two-stage conformer blocks to model both time and frequency dependencies. The decoder then decouples the estimation into a magnitude mask decoder branch to filter out unwanted distortions and a complex refinement branch to further improve the magnitude estimation and implicitly enhance the phase information. Additionally, we include a metric discriminator to alleviate metric mismatch by optimizing the generator with respect to a corresponding evaluation score. Objective and subjective evaluations illustrate that CMGAN is able to show superior performance compared to state-of-the-art methods in three speech enhancement tasks (denoising, dereverberation and super-resolution). For instance, quantitative denoising analysis on Voice Bank+DEMAND dataset indicates that CMGAN outperforms various previous models with a margin, i.e., PESQ of 3.41 and SSNR of 11.10 dB. </p>

show abstract

Icassp 2022 Deep Noise Suppression Challenge

Cited by 95 publications

References 22 publications

FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement

FRCRN: Boosting Feature Representation Using Frequency Recurrence for Monaural Speech Enhancement

Fast Real-time Personalized Speech Enhancement: End-to-End Enhancement Network (E3Net) and Knowledge Distillation

CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement

Contact Info

Product

Resources

About