Workshops & Challenges

Events

Workshops, summer schools, and shared tasks organized and co-organized by GAMMA Lab at major venues in speech, audio, and language research.

2026
Oct 24–29, 2026

SALMA Workshop @ EMNLP 2026

Budapest, Hungary · EMNLP 2026

The Speech and Audio Language Models (SALMA) workshop focuses on LLM-based approaches that treat speech and audio as first-class linguistic modalities rather than secondary inputs — spanning architectural design, training methods, data creation, alignment, and evaluation for large audio-language models across speech, non-speech sound, music, and mixed audio. Co-located with EMNLP 2026.

Workshop website
2025
Apr 6–11, 2025

SALMA Workshop @ ICASSP 2025

Hyderabad, India · ICASSP 2025

The Sound and Language Multimodal Analysis (SALMA) workshop brings together researchers working at the intersection of audio understanding and language models. SALMA explores how large language models can be grounded in acoustic perception — covering audio captioning, question answering, audio-text retrieval, and beyond. Organized by GAMMA Lab members as part of ICASSP 2025.

Workshop website
Jun 9 – Aug 1, 2025

JSALT 2025 Summer Workshop

Brno, Czechia · BUT Faculty of Information Technology

Johns Hopkins Summer Workshop on Language Engineering Technologies (JSALT) is a prestigious six-week research intensive bringing together leading researchers and students to tackle ambitious problems in speech and language. GAMMA Lab members participate in and contribute to collaborative research projects spanning audio understanding, spoken language processing, and multimodal AI at the 2025 edition hosted in Brno.

Workshop website
Oct 30–31, 2025

DCASE 2025 Audio Question Answering Challenge

Barcelona, Spain · DCASE 2025 Workshop

GAMMA Lab co-organizes the Audio Question Answering (AQA) shared task at DCASE 2025, challenging participants to build systems that answer natural language questions about acoustic scenes and events. The challenge is grounded in the MMAU benchmark and evaluates systems on their ability to perform perceptual, semantic, and compositional reasoning about audio content. Workshop held October 30–31 in Barcelona.

Challenge details