# SoundCompass: Navigating Target Sound Extraction With Effective Directional Clue Integration In Complex Acoustic Scenes

arXiv: https://arxiv.org/abs/2509.18561 · HTML: https://mlsanity.com/paper/2509.18561/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 30 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Builds on (most cited first, 22 of 22)

- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [SDR - half-baked or well done?](https://mlsanity.com/paper/1811.02508/) (2018)
- [Understanding and Improving Layer Normalization](https://mlsanity.com/paper/1911.07013/) (2019)
- [Music Source Separation with Band-split RNN](https://mlsanity.com/paper/2209.15174/) (2022)
- [gpuRIR: A Python Library for Room Impulse Response Simulation with GPU Acceleration](https://mlsanity.com/paper/1810.11359/) (2018)
- [Neural Target Speech Extraction: An Overview](https://mlsanity.com/paper/2301.13341/) (2023)
- [Listen to What You Want: Neural Network-based Universal Sound Selector](https://mlsanity.com/paper/2006.05712/) (2020)
- [Semantic Hearing: Programming Acoustic Scenes with Binaural Hearables](https://mlsanity.com/paper/2311.00320/) (2023)
- [Multi-channel Speech Separation Using Spatially Selective Deep Non-linear Filters](https://mlsanity.com/paper/2304.12023/) (2023)
- [ReZero: Region-customizable Sound Extraction](https://mlsanity.com/paper/2308.16892/) (2023)
- [CLAPSep: Leveraging Contrastive Pre-trained Model for Multi-Modal Query-Conditioned Target Sound Extraction](https://mlsanity.com/paper/2402.17455/) (2024)
- [ConceptBeam: Concept Driven Target Speech Extraction](https://mlsanity.com/paper/2207.11964/) (2022)
- [A Generalized Bandsplit Neural Network for Cinematic Audio Source Separation](https://mlsanity.com/paper/2309.02539/) (2023)
- [A Stem-Agnostic Single-Decoder System for Music Source Separation Beyond Four Stems](https://mlsanity.com/paper/2406.18747/) (2024)
- [All Neural Low-latency Directional Speech Extraction](https://mlsanity.com/paper/2407.04879/) (2024)
- [Target Sound Extraction with Variable Cross-modality Clues](https://mlsanity.com/paper/2303.08372/) (2023)
- [Interaural time difference loss for binaural target sound extraction](https://mlsanity.com/paper/2408.00344/) (2024)
- [Language-Queried Target Sound Extraction Without Parallel Training Data](https://mlsanity.com/paper/2409.09398/) (2024)
- [Multichannel-to-Multichannel Target Sound Extraction Using Direction and Timestamp Clues](https://mlsanity.com/paper/2409.12415/) (2024)
- [End-to-End DOA-Guided Speech Extraction in Noisy Multi-Talker Scenarios](https://mlsanity.com/paper/2507.20926/) (2025)
- [Self-Guided Target Sound Extraction and Classification Through Universal Sound Separation Model and Multiple Clues](https://mlsanity.com/paper/2509.13741/) (2025)
- [DeepASA: An Object-Oriented Multi-Purpose Network for Auditory Scene Analysis](https://mlsanity.com/paper/2509.17247/) (2025)

## Abstract (arXiv)

Recent advances in target sound extraction (TSE) utilize directional clues derived from direction of arrival (DoA), which represent an inherent spatial property of sound available in any acoustic scene. However, previous DoA-based methods rely on hand-crafted features or discrete encodings, which lose fine-grained spatial information and limit adaptability. We propose SoundCompass, an effective directional clue integration framework centered on a Spectral Pairwise INteraction (SPIN) module that captures cross-channel spatial correlations in the complex spectrogram domain to preserve full spatial information in multichannel signals. The input feature expressed in terms of spatial correlations is fused with a DoA clue represented as spherical harmonics (SH) encoding. The fusion is carried out across overlapping frequency subbands, inheriting the benefits reported in the previous band-split architectures. We also incorporate the iterative refinement strategy, chain-of-inference (CoI), in the TSE framework, which recursively fuses DoA with sound event activation estimated from the previous inference stage. Experiments demonstrate that SoundCompass, combining SPIN, SH embedding, and CoI, robustly extracts target sources across diverse signal classes and spatial configurations.