L3DAS22 Challenge: Learning 3D Audio Sources in a Real Office Environment

Guizzo, Eric; Marinoni, C.; Pennese, Marco; Ren, Xinlei; Zheng, Xiguang; Zhang, Chen; Masiero, Bruno; Uncini, Aurelio; Comminiello, Danilo

doi:10.48550/arxiv.2202.10372

Cited by 1 publication

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the performance of our model in a robust way, we employ several objective metrics. We compute the F score as suggested by the L3DAS21 Challenge and the error rate (ER) through the location-sensitive detection metrics [11,12]. We build a score from these metrics, which we name location-sensitive detection (LSD score ), defined as…”

Section: Metricsmentioning

confidence: 99%

“…Indeed, while virtual reality (VR) builds spaces different from reallife, augmented reality (AR) expands them through an enlarged user listening experience that is often based on the user immersion into a 3D sound field [5][6][7]. Frequently, this spatial sound field is acquired through First Order Ambisonics (FOA) microphones [8][9][10][11][12], which are arrays of four ideally coincident capsules that decompose the sound field into a combination of spherical harmonics. Due to the strong correlation among the four Ambisonics signals, quaternion neural networks (QNNs) have demonstrated their ability to suitable model these inputs and grasp information coming from each capsule.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dual Quaternion Ambisonics Array for Six-Degree-of-Freedom Acoustic Representation

Grassucci¹,

Mancini²,

Brignone³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Spatial audio methods are gaining a growing interest due to the spread of immersive audio experiences and applications, such as virtual and augmented reality. For these purposes, 3D audio signals are often acquired through arrays of Ambisonics microphones, each comprising four capsules that decompose the sound field in spherical harmonics. In this paper, we propose a dual quaternion representation of the spatial sound field acquired through an array of two First Order Ambisonics (FOA) microphones. The audio signals are encapsulated in a dual quaternion that leverages quaternion algebra properties to exploit correlations among them. This augmented representation with 6 degrees of freedom (6DOF) involves a more accurate coverage of the sound field, resulting in a more precise sound localization and a more immersive audio experience. We evaluate our approach on a sound event localization and detection (SELD) benchmark. We show that our dual quaternion SELD model with temporal convolution blocks (DualQSELD-TCN) achieves better results with respect to real and quaternion-valued baselines thanks to our augmented representation of the sound field. Full code is available at: https://github.com/ispamm/DualQSELD-TCN.

show abstract

Section: Metricsmentioning

confidence: 99%