QC-MI-OVSEL: Benchmarking Query-Conditioned Multi-Instance Open-Vocabulary Sound Event Localization

Spatial audio is central to immersive 360$^\circ$ media, yet existing audio-visual spatialization renders the scene as a whole, leaving no control over an individual source. We introduce Audio-Visual Query-conditioned Target Sound Spatialization (AV-QTSS): given a monaural multi-source mixture, a 360$^\circ$ video and a class query from the 13-class DCASE sound-event ontology, it generates the first-order Ambisonics (FOA) signal of the queried source alone. The omnidirectional mixture contains no explicit spatial cues, so direction has to be inferred from the video, where several objects may sound at once. We propose a query-conditioned dense binding localizer that predicts a spherical probability map over the whole scene, coupled to the renderer so both are optimized jointly. Experiments with one to four concurrent sources show improved localization and spatial accuracy relative to established methods and ablated variants, and the localizer transfers to real 360$^\circ$ recordings after adaptation.

Sound events are entities with semantic identities, locations, and trajectories, but current audio-language models usually reason about clips as global event content. Conversely, sound event localization models track source directions over time but offer limited semantic coverage for language reasoning. To address this gap, we introduce ST-AudioQA, a spatio-temporal audio QA dataset and benchmark built from first-order ambisonic (FOA) renderings of static and moving sound sources. Each scene provides source identity, activity, direction, distance, and motion metadata, enabling dense trajectory supervision and questions about what is sounding, where it is, how it moves, and how sources relate. We further propose ST-Audio Encoder, a time-resolved FOA audio encoder that learns event semantics together with source trajectories, and ST-AudioLM, which connects the audio tokens from the encoder to an LLM for spatio-temporal audio QA. Experiments show that this representation improves the semantic-localization tradeoff and yields stronger reasoning performance than static spatial and localization-oriented baselines.

Understanding dynamic sound sources requires jointly determining what produces a sound, where the source is located, and how it moves over time. Yet existing audio-language models often represent clips as global acoustic events, while vision-language models lack the spatial audio cues needed to localize and track individual sources. To evaluate this missing capability, we introduce ST-OmniQA, a spatio-temporal audio-visual question-answering benchmark built from panoramic videos paired with synchronized first-order Ambisonics (FOA) audio of moving sound sources. It contains 40K videos and 400K question-answer pairs organized into four capability levels covering sound-event recognition, direction of arrival, source distance, motion trajectories, and temporally grounded audio-visual reasoning. Building on this benchmark, we propose ST-Omni-R1, which integrates FOA-derived semantic and trajectory representations with panoramic visual context and is trained through progressive curriculum learning and reasoning-tree reinforcement learning. ST-Omni-R1 achieves 77.83% average semantic accuracy across the four levels, compared with 37.28% for the best evaluated baseline. Results on three public spatial-audio benchmarks further indicate that its learned spatial and motion representations transfer beyond ST-OmniQA.