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 11th Workshop on Detection and Classification of Acoustic Scenes and Events, DCASE 2026, will be held in Boston on 28-29 October. It is co-organized by Bose Corporation, MIT, and Tufts University.

DCASE Workshop 2026 will be co-located with the BioDCASE Workshop (online workshop, date TBD), which focuses on Bio-acoustics, and the SANE Workshop (October 30 at MIT), which is a one-day event gathering researchers and students in speech and audio from the Northeast of the American continent. The SANE Workshop alternates between Boston and New York City every year.

As in previous years, the workshop is organized in conjunction with the DCASE challenge. We aim to bring together researchers from many different universities, research organizations and companies with an interest in the topic, and provide the opportunity for scientific exchange of ideas and opinions.

Topics

We invite submissions on the topics of computational analysis of acoustic scenes and sound events, including but not limited to:

Tasks in computational environmental audio analysis

Environmental audio classification and tagging
Sound event detection and localization
Natural language based audio retrieval
Bio-acoustics
Environmental audio generation
Anomalous sound detection
Audio source separation
Audio Captioning

Multimodal environmental audio analysis and generation

Audio question answering
Audio-language models for acoustic reasoning and scene understanding
Large Audio Language Models (LALMs) for audio, acoustics, and scene grounding
Audio-visual spatial segmentation
Language-guided spatial and embodied audio understanding
Controllable natural language based audio generation
Perception-aligned evaluation of generative audio beyond FID/FAD
Video to audio generation
Multimodal LALM benchmarks
Multimodal representation learning and foundational models

Methods for computational environmental audio analysis

Signal processing and auditory-motivated methods
Machine learning methods: e.g. feature learning, self-supervised learning, foundation modeling for environmental audio
Cross-disciplinary methods involving, e.g., acoustics, biology, psychology, geography, materials science, transports science
Generative modeling
Perceptual analysis and modeling of acoustic environments

Resources, applications, and evaluations of computational environmental-audio analysis

Publicly available datasets: e.g., multichannel dataset, noisy dataset, missing dataset, mismatch device dataset
Publicly available software, taxonomies, and ontologies, evaluation procedures
Benchmark datasets for evaluation
Modeling, simulation, and synthesis of realistic acoustic scenes
Ethics, privacy, responsible research
Applications

We strongly encourage reproducible research with open-source code and open data, though it is not mandatory.

Important notice for challenge participants: Description of systems submitted to the DCASE2026 Challenge is expected to be expanded from the challenge technical report submissions to comply with the format of a scientific paper. This generally means describing the scientific novelty and including more discussions such as ablation studies for additional modules in your method.

Organizers

General Chair

	Shuo Zhang Bose Corp./Tufts University
	Anna Huang Massachusetts Institute of Technology
	Paris Smaragdis Massachusetts Institute of Technology

Technical Program Chairs

	Frederic Font Universitat Pompeu Fabra
	Marko Stamenovic Bose Corp.
	Bashima Islam Worcester Polytechnic Institute
	Mark Cartwright New Jersey Institute of Technology