****https://dcase.community/workshop2026/industry.
📅 When
Tuesday 27 October 2026
08:30 to 17:15, followed by an informal social
📍 Where
314 Main St, 4th Floor, Cambridge, MA 02412
Venue details on the DCASE 2026 website
🎟️ Registration
Included in DCASE 2026 registration
Discounted AIR-only ticket available
| Time | Programme | Format |
|---|---|---|
| 08:30 | Registration and coffee | |
| 09:00 | Welcome from the organisers | |
| 09:15 to 10:30 | Session 1: Listening machines in the real world | |
| 09:15 | Building Ears for Robots: Machine Hearing in the Age of Autonomy | |
| Xuan Zhong (Zoox) | Keynote | |
| 09:45 | Sonaid: Life-Saving Sound Detection on MCU-Class Devices, Tested Live by the Audience | |
| Nicolas Turpault (Sonaid) | Demo | |
| 10:00 | Professionally Recorded and Annotated Non-Speech Audio Corpora for Research & Development | |
| Bradford Swanson, John Sloan, Willy Jones (Pro Sound Effects) | Demo | |
| 10:15 | Designing AI Music Tools With Students: A Balanced Study of Classroom Self-Efficacy and the Participatory Design of an Educational AI Interface | |
| Anish Ghosh (University of Miami) | Demo | |
| 10:30 to 11:00 | ☕ Coffee break and demos | |
| Sonaid, Pro Sound Effects | ||
| 11:00 to 11:45 | Session 2: Simulating acoustics at scale | |
| 11:00 | Generating Large-Scale High-Quality Room-Acoustic Datasets for Sound Event Detection and Localization | |
| Georg Götz, Daniel Gert Nielsen, Vineet Ganju (Treble Technologies) | Tutorial | |
| 11:30 | Scalable Audio Scene Generation with the Treble SDK | |
| Georg Götz, Daniel Gert Nielsen, Vineet Ganju (Treble Technologies) | Demo | |
| 11:45 to 13:00 | Panel: DCASE in the age of rapid AI advancements | |
| Panelists: Noboru Harada (NTT), Jonathan Le Roux (MERL), Kao Kitichotkul (ADI), Marko Stamenovic (Bose), Mrudula Athi (Amazon) | Panel | |
| 13:00 to 14:30 | 🍽️ Lunch and demos | |
| University of Miami, Treble Technologies | ||
| 14:30 to 15:45 | Session 3: From benchmarks to devices | |
| 14:30 | Beyond the Leaderboard: Taking Sound Recognition from DCASE to Millions of Devices | |
| Mateusz Matuszewski (Samsung R&D Institute Poland) | Keynote | |
| 15:00 | Linguistic Sensitivity in Deepfake Speech Detection: What You Read Isn't What You Hear | |
| Thai Le (Indiana University) | Demo | |
| 15:15 | Seiche: An Accessible Tool for Room Acoustics Design | |
| William Chen, Evan Matthews (Independent) | Demo | |
| 15:30 | Cochl.Sense: Real-Time Sound Event Recognition on MCU-Class Devices | |
| Il-Young Jeong (Cochl) | Demo | |
| 15:45 to 16:15 | ☕ Coffee break and demos | |
| Indiana University, Seiche, Cochl | ||
| 16:15 to 17:00 | Session 4: Separation in production | |
| 16:15 | Speech Separation as a Product: Where the Benchmarks Break | |
| Ondřej Cífka, Elio Gruttadauria, Jessica Powell, Hendrik Schreiber, Fabian-Robert Stöter (AudioShake) | Keynote | |
| 17:00 | Closing remarks | |
| 17:15 | 🥂 Informal social |
Keynote
30 min. A technical deep dive and lessons learned from real-world systems.
Tutorial
30 min. A practical introduction to a method, tool or workflow you can reuse.
Demo
15 min on stage, then hands-on at a demo table during one of the breaks.
Nicolas Turpault (Sonaid)
Mikolaj Kegler (Bose)
Sile Yin (Bose)
Tornike Karchkhadze (Apple)
Shuo Zhang (Bose/Tufts)