INTERSPEECH 2021 Acoustic Echo Cancellation Challenge

Cutler, Ross; Saabas, Ando; Pärnamaa, Tanel; Loide, Markus; Sootla, Sten; Purin, Marju; Gamper, Hannes; Braun, Sebastian; Sørensen, Karsten Engsig; Aichner, Robert; Srinivasan, Sriram

doi:10.21437/interspeech.2021-1870

Cited by 41 publications

(27 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The top performing models for Track 1 didn't use additional datasets, while the Track 2 models did. The winning team for Track 1 [25] also won the ICASSP 2022 AEC Challenge [26], and demonstrates a single model can provide excellent AEC, DNS, and WAcc performance. The performance of the personalized DNS track also show excellent performance, greatly exceeding results of our first personalized DNS challenge [2] with the winner [27] providing very good dOVRL with low dSIG and low dWAcc.…”

Section: Resultsmentioning

confidence: 98%

ICASSP 2022 Deep Noise Suppression Challenge

Dubey¹,

Gopal²,

Cutler³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The Deep Noise Suppression (DNS) challenge is designed to foster innovation in the area of noise suppression to achieve superior perceptual speech quality. This is the 4th DNS challenge, with the previous editions held at INTERSPEECH 2020 [1], ICASSP 2021 [2], and INTERSPEECH 2021 [3]. We open-source datasets and test sets for researchers to train their deep noise suppression models, as well as a subjective evaluation framework based on ITU-T P.835 to rate and rank-order the challenge entries. We provide access to DNS-MOS P.835 and word accuracy (WAcc) APIs to challenge participants to help with iterative model improvements. In this challenge, we introduced the following changes: (i) Included mobile device scenarios in the blind test set; (ii) Included a personalized noise suppression track with baseline; (iii) Added WAcc as an objective metric; (iv) Included DNSMOS P.835; (v) Made the training datasets and test sets fullband (48 kHz). We use an average of WAcc and subjective scores P.835 SIG, BAK, and OVRL to get the final score for ranking the DNS models. We believe that as a research community, we still have a long way to go in achieving excellent speech quality in challenging noisy real-world scenarios.

show abstract

Section: Resultsmentioning

confidence: 98%

ICASSP 2022 Deep Noise Suppression Challenge

Dubey¹,

Gopal²,

Cutler³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We experiment on the AEC-challenge data [13] to validate the proposed method. In order to train the network, four types of signals need to be prepared: near-end speech, background noise, far-end speech and corresponding echo signal.…”

Section: Datasetmentioning

confidence: 99%

“…For on-the-fly data generation, RIRs are only used to convolve with near-end speech s(n) with 50% probability. The far-end speech x(n) and echo signal d(n) are either already reverberated or real recordings in different rooms [13], so there is no need for reverberation.…”

Section: Data Augmentationmentioning

confidence: 99%

“…Fazel et al [10] designed a deep contextual-attention module with frequency domain NLMS to adaptively estimate features of the near-end speech. Wang et al [11] and Valin et al [12] have also achieved competitive results in the recent AEC-challenge [13]. Zhang and Wang [14] formulated AEC as a supervised speech separation problem, where a bidirectional long-short term memory (BLSTM) network was adopted to predict a mask for magnitude of microphone signal.…”

Section: Introductionmentioning

confidence: 99%

“…The superior performance is mainly attributed to the effective use of the phase information. Moreover, complex domain based methods have achieved overall better subjective listening performance in the Deep Noise Suppression (DNS) Challenge [13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

F-T-LSTM based Complex Network for Joint Acoustic Echo Cancellation and Speech Enhancement

Zhang,

Kong,

et al. 2021

Preprint

View full text Add to dashboard Cite

With the increasing demand for audio communication and online conference, ensuring the robustness of Acoustic Echo Cancellation (AEC) under the complicated acoustic scenario including noise, reverberation and nonlinear distortion has become a top issue. Although there have been some traditional methods that consider nonlinear distortion, they are still inefficient for echo suppression and the performance will be attenuated when noise is present. In this paper, we present a real-time AEC approach using complex neural network to better modeling the important phase information and frequency-time-LSTMs (F-T-LSTM), which scan both frequency and time axis, for better temporal modeling. Moreover, we utilize modified SI-SNR as cost function to make the model to have better echo cancellation and noise suppression (NS) performance. With only 1.4M parameters, the proposed approach outperforms the AECchallenge baseline by 0.27 in terms of Mean Opinion Score (MOS).

show abstract

Hybrid metaheuristic method of ABC kernel filtering for nonlinear acoustic echo cancellation

Diana¹,

Carline²

2023

Applied Acoustics

View full text Add to dashboard Cite

INTERSPEECH 2021 Acoustic Echo Cancellation Challenge

Cited by 41 publications

References 8 publications

ICASSP 2022 Deep Noise Suppression Challenge

ICASSP 2022 Deep Noise Suppression Challenge

F-T-LSTM based Complex Network for Joint Acoustic Echo Cancellation and Speech Enhancement

Hybrid metaheuristic method of ABC kernel filtering for nonlinear acoustic echo cancellation

Contact Info

Product

Resources

About