# Assessment of Self-Attention on Learned Features For Sound Event Localization and Detection

arXiv: https://arxiv.org/abs/2107.09388 · HTML: https://mlsanity.com/paper/2107.09388/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jul 2021.

## Built on by (most cited first, 7 of 7)

- [A Survey of Sound Source Localization with Deep Learning Methods](https://mlsanity.com/paper/2109.03465/) (2021)
- [STARSS23: An Audio-Visual Dataset of Spatial Recordings of Real Scenes with Spatiotemporal Annotations of Sound Events](https://mlsanity.com/paper/2306.09126/) (2023)
- [Stereo Sound Event Localization and Detection with Onscreen/offscreen Classification](https://mlsanity.com/paper/2507.12042/) (2025)
- [Enhanced Sound Event Localization and Detection in Real 360-degree audio-visual soundscapes](https://mlsanity.com/paper/2401.17129/) (2024)
- [Attention-Based Methods For Audio Question Answering](https://mlsanity.com/paper/2305.19769/) (2023)
- [Exploring Self-Supervised Contrastive Learning of Spatial Sound Event Representation](https://mlsanity.com/paper/2309.15938/) (2023)
- [Latent Acoustic Mapping for Direction of Arrival Estimation: A Self-Supervised Approach](https://mlsanity.com/paper/2507.07066/) (2025)

## Abstract (arXiv)

Joint sound event localization and detection (SELD) is an emerging audio signal processing task adding spatial dimensions to acoustic scene analysis and sound event detection. A popular approach to modeling SELD jointly is using convolutional recurrent neural network (CRNN) models, where CNNs learn high-level features from multi-channel audio input and the RNNs learn temporal relationships from these high-level features. However, RNNs have some drawbacks, such as a limited capability to model long temporal dependencies and slow training and inference times due to their sequential processing nature. Recently, a few SELD studies used multi-head self-attention (MHSA), among other innovations in their models. MHSA and the related transformer networks have shown state-of-the-art performance in various domains. While they can model long temporal dependencies, they can also be parallelized efficiently. In this paper, we study in detail the effect of MHSA on the SELD task. Specifically, we examined the effects of replacing the RNN blocks with self-attention layers. We studied the influence of stacking multiple self-attention blocks, using multiple attention heads in each self-attention block, and the effect of position embeddings and layer normalization. Evaluation on the DCASE 2021 SELD (task 3) development data set shows a significant improvement in all employed metrics compared to the baseline CRNN accompanying the task.