# Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learning

arXiv: https://arxiv.org/abs/2308.09514 · HTML: https://mlsanity.com/paper/2308.09514/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 41 references that are in the map.
- First posted on arXiv 18 Aug 2023.

## Builds on (most cited first, 18 of 18)

- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [Sound Event Localization and Detection of Overlapping Sources Using Convolutional Recurrent Neural Networks](https://mlsanity.com/paper/1807.00129/) (2018)
- [Sudo rm -rf: Efficient Networks for Universal Audio Source Separation](https://mlsanity.com/paper/2007.06833/) (2020)
- [Diversity in Machine Learning](https://mlsanity.com/paper/1807.01477/) (2018)
- [Interspeech 2021 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2101.01902/) (2021)
- [A Survey of Sound Source Localization with Deep Learning Methods](https://mlsanity.com/paper/2109.03465/) (2021)
- [Building and Evaluation of a Real Room Impulse Response Dataset](https://mlsanity.com/paper/1811.06795/) (2018)
- [EasyCom: An Augmented Reality Dataset to Support Algorithms for Easy Communication in Noisy Environments](https://mlsanity.com/paper/2107.04174/) (2021)
- [A Dataset of Dynamic Reverberant Sound Scenes with Directional Interferers for Sound Event Localization and Detection](https://mlsanity.com/paper/2106.06999/) (2021)
- [dEchorate: a Calibrated Room Impulse Response Database for Echo-aware Signal Processing](https://mlsanity.com/paper/2104.13168/) (2021)
- [Dataset of Spatial Room Impulse Responses in a Variable Acoustics Room for Six Degrees-of-Freedom Rendering and Analysis](https://mlsanity.com/paper/2111.11882/) (2021)
- [PILOT: Introducing Transformers for Probabilistic Sound Event Localization](https://mlsanity.com/paper/2106.03903/) (2021)
- [The DIRHA-English corpus and related tasks for distant-speech recognition in domestic environments](https://mlsanity.com/paper/1710.02560/) (2017)
- [Blind Room Parameter Estimation Using Multiple-Multichannel Speech Recordings](https://mlsanity.com/paper/2107.13832/) (2021)
- [A data acquisition setup for data driven acoustic design](https://mlsanity.com/paper/2109.12014/) (2021)
- [BIRD: Big Impulse Response Dataset](https://mlsanity.com/paper/2010.09930/) (2020)
- [Deep Neural Network based Distance Estimation for Geometry Calibration in Acoustic Sensor Networks](https://mlsanity.com/paper/2006.13769/) (2020)

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

- [Learning Spatially-Aware Language and Audio Embeddings](https://mlsanity.com/paper/2409.11369/) (2024)
- [ImmerseDiffusion: A Generative Spatial Audio Latent Diffusion Model](https://mlsanity.com/paper/2410.14945/) (2024)
- [Novel-View Acoustic Synthesis from 3D Reconstructed Rooms](https://mlsanity.com/paper/2310.15130/) (2023)
- [AudioMotionBench: Evaluating Auditory Motion Perception in Audio LLMs](https://mlsanity.com/paper/2511.13273/) (2025)
- [Blind Estimation of Sub-band Acoustic Parameters from Ambisonics Recordings using Spectro-Spatial Covariance Features](https://mlsanity.com/paper/2411.03172/) (2024)
- [RIPPLE: Generating Multi-Channel Phase, Not Recovering It](https://mlsanity.com/paper/2607.27775/) (2026)
- [SonicWorld: Source-Centric Latent Dynamics for Spatial Audio Reasoning](https://mlsanity.com/paper/2610.17131/) (2026)

## Abstract (arXiv)

We present Spatial LibriSpeech, a spatial audio dataset with over 650 hours of 19-channel audio, first-order ambisonics, and optional distractor noise. Spatial LibriSpeech is designed for machine learning model training, and it includes labels for source position, speaking direction, room acoustics and geometry. Spatial LibriSpeech is generated by augmenting LibriSpeech samples with 200k+ simulated acoustic conditions across 8k+ synthetic rooms. To demonstrate the utility of our dataset, we train models on four spatial audio tasks, resulting in a median absolute error of 6.60deg on 3D source localization, 0.43m on distance, 90.66ms on T30, and 2.74dB on DRR estimation. We show that the same models generalize well to widely-used evaluation datasets, e.g., obtaining a median absolute error of 12.43deg on 3D source localization on TUT Sound Events 2018, and 157.32ms on T30 estimation on ACE Challenge.