cs.SDNov 6, 2026

SAS: Scene-Annotated AudioSet Strong for Joint Scene-Event Analysis

Authors: Michel Olvera, Paraskevas Stamatiadis, Romain Serizel, Slim Essid, Gaël Richard

Organizations: S2A, LTCI, IDS · MULTISPEECH · IDS, S2A, LTCI · S2A, IDS

Abstract

Context is central to human auditory perception, as listeners interpret sounds within their acoustic environment to resolve ambiguity. However, large-scale datasets that jointly capture sound events and acoustic scenes remain limited. We address this by augmenting AudioSet Strong with clip-level scene annotations under two taxonomies for joint sceneevent modeling. A corpus-scale analysis shows that scene identity is better captured by events distinctive to a scene than by those that merely occur frequently. Reweighting event co-occurrences toward scene-specific associations produces representations that better align with human-derived taxonomies, while also exposing scenes where the acoustic and semantic views diverge. This geometric divergence predicts classification errors, while supervised validation confirms that the proposed scene annotations provide meaningful complementary information for developing scalable context-aware machine listening systems.

Figures & tables

Explore similar work

CardsList
  1. CARES: A Controlled Synthetic Benchmark of Speaker Reactions to Sound

    Oct 7, 2026Marcel Gibier, Thomas Thebaud, Olivier Boëffard +1Audio UnderstandingAudio Captioning

  2. From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models

    Jun 24, 2026Pengfei Zhang, Hoang H Nguyen, Kazi Shaharair Sharif +6Audio-Language Model EvaluationAudio Understanding

  3. Logbook: Extremely Long-form Audio Event Understanding

    Oct 5, 2026Kwanghee Choi, Suwon Shon, Dmitriy Serdyuk +7Audio-Language Model EvaluationAudio Understanding