Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learning

Miguel Sarabia, Elena Menyaylenko, Alessandro Toso, Skyler Seto, Zakaria Aldeneh +5 more
8/18/2023
cs.SDcs.AIcs.LG

Official code: apple/ml-spatial-librispeech (129 stars, last commit 2023-10-25).

Abstract

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 6.60° 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 12.43° on 3D source localization on TUT Sound Events 2018, and 157.32ms on T30 estimation on ACE Challenge.

View on arXivView PDF

Code Implementations(1)

apple/ml-spatial-librispeechOfficial100%

A large synthetic dataset of spatial audio with multiple labels

12911Aug 18, 20232 years agoNOASSERTION
machine-learningspatial-audio

Similar Papers

BAT: Learning to Reason about Spatial Sounds with Large Language Models

85% match1 repo902026

Learning Spatially-Aware Language and Audio Embeddings

85% match2024

Sci-Phi: A Large Language Model Spatial Audio Descriptor

85% match2025

The World is Not Mono: Enabling Spatial Understanding in Large Audio-Language Models

84% match2026

DSpAST: Disentangled Representations for Spatial Audio Reasoning with Large Language Models

83% match2 repos172025

JAEGER: Joint 3D Audio-Visual Grounding and Reasoning in Simulated Physical Environments

83% match2026

Cite this paper

@article{sarabia2023spatial,
  title  = {Spatial LibriSpeech: An Augmented Dataset for Spatial Audio Learning},
  author = {Miguel Sarabia and Elena Menyaylenko and Alessandro Toso and Skyler Seto and Zakaria Aldeneh and Shadi Pirhosseinloo and Luca Zappella and Barry-John Theobald and Nicholas Apostoloff and Jonathan Sheaffer},
  year   = {2023},
  eprint = {2308.09514},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2308.09514v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2308.09514.svg)](https://codeofpaper.com/paper/2308.09514)

Discussion