INTERSPEECH 2022 Audio Deep Packet Loss Concealment Challenge

Diener, Lorenz; Sootla, Sten; Branets, Solomiya; Saabas, Ando; Aichner, Robert; Cutler, Ross

doi:10.21437/interspeech.2022-10829

Cited by 21 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While this challenge focuses on improving speech quality by reducing background noise (improving BAK) and reverberation which improves OVRL, other related challenges target echo cancellation [41]- [44], packet loss concealment [45], and general speech signal improvements [46]. The ICASSP 2023 Speech Signal Improvement challenge [46] provides test sets with various types of SIG regressions such as poor-quality microphones and speech enhancement.…”

Section: Related Workmentioning

confidence: 99%

ICASSP 2023 Deep Noise Suppression Challenge

Dubey,

Aazami,

Gopal

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

The ICASSP 2023 Deep Noise Suppression (DNS) Challenge marks the fifth edition of the DNS challenge series. DNS challenges were organized from 2019 to 2023 to foster research in the field of DNS. Previous DNS challenges were held at INTERSPEECH 2020, ICASSP 2021, INTERSPEECH 2021, and ICASSP 2022 This challenge aims to advance models capable of jointly addressing denoising, dereverberation, and interfering talker suppression, with separate tracks focusing on headset and speakerphone scenarios. The challenge facilitates personalized deep noise suppression by providing accompanying enrollment clips for each test clip, each containing the primary talker only, which can be used to compute a speaker identity feature and disentangle primary and interfering speech. While the majority of models submitted to the challenge were personalized, the same teams emerged as the winners in both tracks. The best models demonstrated improvements of 0.145 and 0.141 in the challenge's score, respectively, when compared to the noisy blind test set. We present additional analysis and draw comparisons to previous challenges.

show abstract

Section: Related Workmentioning

confidence: 99%

ICASSP 2023 Deep Noise Suppression Challenge

Dubey,

Aazami,

Gopal

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Noisiness Deep Noise Suppression [21], [22], [3], [2], [4] Coloration None Discontinuity Packet Loss Concealment [23] Loudness None Reverberation REVERB [24] Echo AEC Challenge [25], [26], [27], [28] • A overlap-save processing has a buffering latency corresponding to the frame size. • A time-domain convolution with stride 1 introduces a buffering latency of 1 sample.…”

Section: Area Related Challengementioning

confidence: 99%

ICASSP 2023 Speech Signal Improvement Challenge

Cutler,

Saabas,

Naderi

et al. 2024

IEEE Open J. Signal Process.

View full text Add to dashboard Cite

The ICASSP 2023 Speech Signal Improvement Challenge is intended to stimulate research in the area of improving the speech signal quality in communication systems. The speech signal quality can be measured with SIG in ITU-T P.835 and is still a top issue in audio communication and conferencing systems. For example, in the ICASSP 2022 Deep Noise Suppression challenge, the improvement in the background and overall quality is impressive, but the improvement in the speech signal is not statistically significant. To improve the speech signal the following speech impairment areas must be addressed: coloration, discontinuity, loudness, reverberation, and noise. A training and test set was provided for the challenge, and the winners were determined using an extended crowdsourced implementation of ITU-T P.804's listening phase. The results show significant improvement was made across all measured dimensions of speech quality.

show abstract

“…The dataset for the ICASSP 2024 challenge is built upon the same framework as before, leveraging real-world packet loss patterns combined with data that is either in the public domain (conversational speech, sourced from the LibriVox Community Podcast) 1 or was collected by us explicitly for use in challenges (read speech), allowing us to have a realistic dataset while avoiding the potential for privacy issues. Audio segments were selected by filtering using DNSMOS [3] and manual inspection to avoid very noisy base audio clips, and were cut to 10 to 15 seconds of length using the WebRTC Voice Activity Detection to avoid cutting off parts of words.…”

Section: Dataset Constructionmentioning

confidence: 99%

“…• 0-120 ms burst: 20 per loss bracket, 100 total Additionally, we include a total of 200 traces also used in the 2022 PLC Challenge to allow for limited comparability. Further details about the dataset creation procedure and data sourcing can be found in our previous work [1]. On Oct. 11, 2023, we first released a validation set constructed in this way, followed by a blind set with no references on Dec. 1, 2023.…”

Section: Dataset Constructionmentioning

confidence: 99%

“…As algorithms and hardware have advanced, it is now possible to perform PLC using machine learning rather than basic digital signal processing, with potential for vast quality improvements. In the PLC Challenge held at INTERSPEECH 2022, we for the first time brought together researchers working on the topic to compare approaches on a common dataset, with many interesting approaches and results [1].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation