The tasks for DCASE challenge 2025 have been announced.
dcase.community/articles/cha...
Stay tuned for more details.
dcase.community/articles/cha...
Stay tuned for more details.
Challenge tasks for DCASE2025 - DCASE
The DCASE Steering Group has reviewed the task proposals...
dcase.community
December 10, 2024 at 10:13 AM
The tasks for DCASE challenge 2025 have been announced.
dcase.community/articles/cha...
Stay tuned for more details.
dcase.community/articles/cha...
Stay tuned for more details.
📢 The short description of the tasks is now available on the website 👇
dcase.community/challenge2025/
DCASE2025 Challenge - DCASE
Introduction Sounds carry a large amount of information...
dcase.community
January 20, 2025 at 2:19 PM
📢 The short description of the tasks is now available on the website 👇
dcase.community/challenge2025/
DCASE evaluation phase is now open! You have until June 15th 23:59 AoE to submit your systems!
dcase.community/challenge2025/
dcase.community/challenge2025/
DCASE2025 Challenge - DCASE
Challenge status Task Task description Development dataset...
dcase.community
June 2, 2025 at 2:13 PM
DCASE evaluation phase is now open! You have until June 15th 23:59 AoE to submit your systems!
dcase.community/challenge2025/
dcase.community/challenge2025/
You can watch the livestream of #BioDCASE - here: dcase.community/workshop2025...
DCASE2025 Live Stream - DCASE
Live Stream Most of DCASE2025 Workshop sessions happening...
dcase.community
October 29, 2025 at 8:45 AM
You can watch the livestream of #BioDCASE - here: dcase.community/workshop2025...
📢 DCASE 2025 challenge is now officially launched!
🎉
You'll find more info about the tasks of the challenge website 👇
dcase.community/challenge2025/
🎉
You'll find more info about the tasks of the challenge website 👇
dcase.community/challenge2025/
DCASE2025 Challenge - DCASE
Challenge status Task Task description Development dataset...
dcase.community
April 2, 2025 at 8:57 AM
📢 DCASE 2025 challenge is now officially launched!
🎉
You'll find more info about the tasks of the challenge website 👇
dcase.community/challenge2025/
🎉
You'll find more info about the tasks of the challenge website 👇
dcase.community/challenge2025/
This year the DCASE Challenge received 275 submission entries from 86 teams. Task-wise details at dcase.community/articles/dca...
We start the evaluation process, aiming to publish the results on 30th of June. Stay tuned!
@dcase-challenge.bsky.social
We start the evaluation process, aiming to publish the results on 30th of June. Stay tuned!
@dcase-challenge.bsky.social
DCASE2025 Challenge received 321 submission entries - DCASE
DCASE2025 Challenge received 275 submission entries
dcase.community
June 19, 2025 at 12:44 PM
This year the DCASE Challenge received 275 submission entries from 86 teams. Task-wise details at dcase.community/articles/dca...
We start the evaluation process, aiming to publish the results on 30th of June. Stay tuned!
@dcase-challenge.bsky.social
We start the evaluation process, aiming to publish the results on 30th of June. Stay tuned!
@dcase-challenge.bsky.social
Excited to be travelling to BioDCASE this week! https://biodcase.github.io/workshop2025/ Our full-day workshop on #bioacoustics data challenges, in #barcelona, hosted by #dcase2025! Looking forward to learning about people's work on AI for nature and the ways forward.
BioDCASE Workshop - BioDCASE
Join us for the BioDCASE Workshop held in Barcelona, Spain on the 29th of October! The workshop will be held at the Campus del Poblenou of Universitat Pompeu Fabra. The BioDCASE workshop will be hosted the day before the DCASE workshop on the 30-31st of October at the same venue …
biodcase.github.io
October 27, 2025 at 11:49 AM
Excited to be travelling to BioDCASE this week! https://biodcase.github.io/workshop2025/ Our full-day workshop on #bioacoustics data challenges, in #barcelona, hosted by #dcase2025! Looking forward to learning about people's work on AI for nature and the ways forward.
Launching: BioDCASE - the Bioacoustics Data Challenge! https://biodcase.github.io/ #DCASE #dcase2025 #dclde #bioacoustics #ai4good
BioDCASE
About BioDCASE - Evaluation & Benchmarking in Automated Bioacoustics BioDCASE is a new initiative focused on advancing research in computational bioacoustics through annual challenges and workshops, modelled after the successful format of the DCASE challenges. By providing benchmark datasets and tasks specific to bioacoustic applications, BioDCASE aims to foster innovation and collaboration …
biodcase.github.io
February 14, 2025 at 1:27 PM
Launching: BioDCASE - the Bioacoustics Data Challenge! https://biodcase.github.io/ #DCASE #dcase2025 #dclde #bioacoustics #ai4good
You can watch the live stream of #biodcase! Here: https://dcase.community/workshop2025/live-stream
DCASE2025 Live Stream - DCASE
Live Stream Most of DCASE2025 Workshop sessions happening at the Auditorium of UPF will be live-streamed at this page. Check the schedule for details about the different sessions. video#video-player1134442278 { position: relative; display: block; width: 100%; margin-bottom: 2em; } @media (max-width: 978px) { video#video-player1134442278 { display: block !important; } } if (IVSPlayer.isPlayerSupported) { const …
dcase.community
October 29, 2025 at 8:47 AM
You can watch the live stream of #biodcase! Here: https://dcase.community/workshop2025/live-stream
You can watch the live stream of #biodcase! Here: https://dcase.community/workshop2025/live-stream
DCASE2025 Live Stream - DCASE
Live Stream Most of DCASE2025 Workshop sessions happening at the Auditorium of UPF will be live-streamed at this page. Check the schedule for details about the different sessions. video#video-player1134442278 { position: relative; display: block; width: 100%; margin-bottom: 2em; } @media (max-width: 978px) { video#video-player1134442278 { display: block !important; } } if (IVSPlayer.isPlayerSupported) { const …
dcase.community
October 29, 2025 at 8:47 AM
You can watch the live stream of #biodcase! Here: https://dcase.community/workshop2025/live-stream
The MTG organizes the next Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE2025) at @upf.edu (October 30-31) + satellite events: BioDCASE (Oct 29) and Freesound 20 anniversary celebration day (Oct 28). The full program will be available soon www.upf.edu/web/mtg/home...
The MTG organizes the next Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE2025) - MTG - Music Technology Group - UPF
It will take place at Universitat Pompeu Fabra, Barcelona, in October 2025
www.upf.edu
September 12, 2025 at 12:10 PM
The MTG organizes the next Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE2025) at @upf.edu (October 30-31) + satellite events: BioDCASE (Oct 29) and Freesound 20 anniversary celebration day (Oct 28). The full program will be available soon www.upf.edu/web/mtg/home...
Stereo SELD system uses pre-trained PSELDnet and bidirectional Mamba, replacing Conformer with BiMamba, improving performance and reducing computing resources on the DCASE2025 Task 3 dataset.
Enhancing Stereo Sound Event Detection with BiMamba and Pretrained PSELDnet
Wenmiao Gao, Han Yin
arxiv.org
July 15, 2025 at 9:22 AM
Stereo SELD system uses pre-trained PSELDnet and bidirectional Mamba, replacing Conformer with BiMamba, improving performance and reducing computing resources on the DCASE2025 Task 3 dataset.
Call for papers! #DCASE 2025, will be held in Barcelona on 30-31 October (plus #biodcase on 29 Oct). We invite 4-page papers, to be submitted by 11th July. Full details: https://dcase.community/workshop2025/call-for-papers #cfp #acoustics #research #machinelistening
DCASE2025 Workshop, Call for Papers - DCASE
The workshop aims to provide a venue for researchers working on computational analysis of sound events and scene analysis to present and discuss their results. The 10th Workshop on Detection and Classification of Acoustic Scenes and Events, DCASE 2025, will be held in Barcelona on 30-31 October. Moreover, a satellite …
dcase.community
June 20, 2025 at 12:05 PM
Call for papers! #DCASE 2025, will be held in Barcelona on 30-31 October (plus #biodcase on 29 Oct). We invite 4-page papers, to be submitted by 11th July. Full details: https://dcase.community/workshop2025/call-for-papers #cfp #acoustics #research #machinelistening
Hyun Jun Kim, Hyeong Yong Choi, Changwon Lim: AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval https://arxiv.org/abs/2509.16649 https://arxiv.org/pdf/2509.16649 https://arxiv.org/html/2509.16649
September 23, 2025 at 6:35 AM
Hyun Jun Kim, Hyeong Yong Choi, Changwon Lim: AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval https://arxiv.org/abs/2509.16649 https://arxiv.org/pdf/2509.16649 https://arxiv.org/html/2509.16649
The DCASE 2025 Challenge introduced Task 2, a first‑shot unsupervised anomalous sound detection problem, drawing 119 submissions from 35 international research teams. Read more: https://getnews.me/dcase-2025-challenge-introduces-first-shot-unsupervised-sound-detection/ #dcase2025 #anomaloussound
September 29, 2025 at 10:45 PM
The DCASE 2025 Challenge introduced Task 2, a first‑shot unsupervised anomalous sound detection problem, drawing 119 submissions from 35 international research teams. Read more: https://getnews.me/dcase-2025-challenge-introduces-first-shot-unsupervised-sound-detection/ #dcase2025 #anomaloussound
AISTAT's dual‑encoder system for DCASE 2025 Task 6, using contrastive learning, achieved a mean average precision@16 of 46.62, rising to 48.83 with a four‑model ensemble. Read more: https://getnews.me/aistat-lab-advances-language-based-audio-retrieval-for-dcase-2025/ #dcase2025 #aistat
September 24, 2025 at 12:04 PM
AISTAT's dual‑encoder system for DCASE 2025 Task 6, using contrastive learning, achieved a mean average precision@16 of 46.62, rising to 48.83 with a four‑model ensemble. Read more: https://getnews.me/aistat-lab-advances-language-based-audio-retrieval-for-dcase-2025/ #dcase2025 #aistat
A new study shows that attribute-aware pre-training with pseudo-attribute labels boosts anomalous sound detection, achieving higher accuracy on the DCASE 2025 Challenge dataset. https://getnews.me/attribute-aware-pre-training-boosts-anomalous-sound-detection/ #anomaloussounddetection #dcase2025
September 23, 2025 at 2:23 AM
A new study shows that attribute-aware pre-training with pseudo-attribute labels boosts anomalous sound detection, achieving higher accuracy on the DCASE 2025 Challenge dataset. https://getnews.me/attribute-aware-pre-training-boosts-anomalous-sound-detection/ #anomaloussounddetection #dcase2025
Dual encoder architecture employs contrastive learning; uses data augmentation techniques, including back-translation and LLM mix. A clustering auxillary classification task for further finetuning achieves mAP@16 of 48.83 on the Clotho development test split.
AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval
Hyun Jun Kim, Hyeong Yong Choi, Changwon Lim
arxiv.org
September 23, 2025 at 8:03 AM
Dual encoder architecture employs contrastive learning; uses data augmentation techniques, including back-translation and LLM mix. A clustering auxillary classification task for further finetuning achieves mAP@16 of 48.83 on the Clotho development test split.
Stereo SELD task in DCASE2025 challenge uses stereo audio and introduces onscreen/offscreen classification; baseline system processes stereo audio/video, incorporating new accuracy metric.
Stereo Sound Event Localization and Detection with Onscreen/offscreen Classification
Kazuki Shimada, Archontis Politis, Iran R. Roman, Parthasaarathy Sudarsanam, David Diaz-Guerra, Ruchi Pandey, Kengo Uchida, Yuichiro Koyama, Naoya Takahashi, Takashi Shibuya, Shusuke Takahashi, Tuomas Virtanen, Yuki Mitsufuji
arxiv.org
July 17, 2025 at 9:02 AM
Stereo SELD task in DCASE2025 challenge uses stereo audio and introduces onscreen/offscreen classification; baseline system processes stereo audio/video, incorporating new accuracy metric.
A stereo SELD system replaces the Conformer module in PSELDnet with a BiMamba module and introduces asymmetric convolutions, improving performance and reducing computational complexity on the DCASE2025 Task 3 dataset.
Stereo sound event localization and detection based on PSELDnet pretraining and BiMamba sequence modeling
Wenmiao Gao, Yang Xiao
arxiv.org
June 17, 2025 at 11:36 AM
A stereo SELD system replaces the Conformer module in PSELDnet with a BiMamba module and introduces asymmetric convolutions, improving performance and reducing computational complexity on the DCASE2025 Task 3 dataset.