Lorenzo Picinali

Aidan O. T. Hogg, Mads Jenkins, He Liu et al.

An individualised head-related transfer function (HRTF) is very important for creating realistic virtual reality (VR) and augmented reality (AR) environments. However, acoustically measuring high-quality HRTFs requires expensive equipment and an acoustic lab setting. To overcome these limitations and to make this measurement more efficient HRTF upsampling has been exploited in the past where a high-resolution HRTF is created from a low-resolution one. This paper demonstrates how generative adversarial networks (GANs) can be applied to HRTF upsampling. We propose a novel approach that transforms the HRTF data for direct use with a convolutional super-resolution generative adversarial network (SRGAN). This new approach is benchmarked against three baselines: barycentric upsampling, spherical harmonic (SH) upsampling and an HRTF selection approach. Experimental results show that the proposed method outperforms all three baselines in terms of log-spectral distortion (LSD) and localisation performance using perceptual models when the input HRTF is sparse (less than 20 measured positions).

Johan Pauwels, Lorenzo Picinali

As spatial audio is enjoying a surge in popularity, data-driven machine learning techniques that have been proven successful in other domains are increasingly used to process head-related transfer function measurements. However, these techniques require much data, whereas the existing datasets are ranging from tens to the low hundreds of datapoints. It therefore becomes attractive to combine multiple of these datasets, although they are measured under different conditions. In this paper, we first establish the common ground between a number of datasets, then we investigate potential pitfalls of mixing datasets. We perform a simple experiment to test the relevance of the remaining differences between datasets when applying machine learning techniques. Finally, we pinpoint the most relevant differences.

Xuyi Hu, Jian Li, Lorenzo Picinali et al.

The demand for realistic virtual immersive audio continues to grow, with Head-Related Transfer Functions (HRTFs) playing a key role. HRTFs capture how sound reaches our ears, reflecting unique anatomical features and enhancing spatial perception. It has been shown that personalized HRTFs improve localization accuracy, but their measurement remains time-consuming and requires a noise-free environment. Although machine learning has been shown to reduce the required measurement points and, thus, the measurement time, a controlled environment is still necessary. This paper proposes a method to address this constraint by presenting a novel technique that can upsample sparse, noisy HRTF measurements. The proposed approach combines an HRTF Denoisy U-Net for denoising and an Autoencoding Generative Adversarial Network (AE-GAN) for upsampling from three measurement points. The proposed method achieves a log-spectral distortion (LSD) error of 5.41 dB and a cosine similarity loss of 0.0070, demonstrating the method's effectiveness in HRTF upsampling.

Xuyi Hu, Jian Li, Shaojie Zhang et al.

Personalized Head-Related Transfer Functions (HRTFs) are starting to be introduced in many commercial immersive audio applications and are crucial for realistic spatial audio rendering. However, one of the main hesitations regarding their introduction is that creating personalized HRTFs is impractical at scale due to the complexities of the HRTF measurement process. To mitigate this drawback, HRTF spatial upsampling has been proposed with the aim of reducing measurements required. While prior work has seen success with different machine learning (ML) approaches, these models often struggle with long-range spatial consistency and generalization at high upsampling factors. In this paper, we propose a novel transformer-based architecture for HRTF upsampling, leveraging the attention mechanism to better capture spatial correlations across the HRTF sphere. Working in the spherical harmonic (SH) domain, our model learns to reconstruct high-resolution HRTFs from sparse input measurements with significantly improved accuracy. To enhance spatial coherence, we introduce a neighbor dissimilarity loss that promotes magnitude smoothness, yielding more realistic upsampling. We evaluate our method using both perceptual localization models and objective spectral distortion metrics. Experiments show that our model surpasses leading methods by a substantial margin in generating realistic, high-fidelity HRTFs.

Marco Comunità, Andrea Gerino, Veranika Lim et al.

PlugSonic is a suite of web- and mobile-based applications for the curation and experience of binaural interactive soundscapes and sonic narratives. It was developed as part of the PLUGGY EU project (Pluggable Social Platform for Heritage Awareness and Participation) and consists of two main applications: PlugSonic Sample, to edit and apply audio effects, and PlugSonic Soundscape, to create and experience binaural soundscapes. The audio processing within PlugSonic is based on the Web Audio API and the 3D Tune-In Toolkit, while the exploration of soundscapes in a physical space is obtained using Apple's ARKit. In this paper we present the design choices, the user involvement processes and the implementation details. The main goal of PlugSonic is technology democratisation; PlugSonic users - whether institutions or citizens - are all given the instruments needed to create, process and experience 3D soundscapes and sonic narrative; without the need for specific devices, external tools (software and/or hardware), specialised knowledge or custom development. The evaluation, which was conducted with inexperienced users on three tasks - creation, curation and experience - demonstrates how PlugSonic is indeed a simple, effective, yet powerful tool.

Sergio Mascetti, Lorenzo Picinali, Andrea Gerino et al.

In the last years several solutions were proposed to support people with visual impairments or blindness during road crossing. These solutions focus on computer vision techniques for recognizing pedestrian crosswalks and computing their relative position from the user. Instead, this contribution addresses a different problem; the design of an auditory interface that can effectively guide the user during road crossing. Two original auditory guiding modes based on data sonification are presented and compared with a guiding mode based on speech messages. Experimental evaluation shows that there is no guiding mode that is best suited for all test subjects. The average time to align and cross is not significantly different among the three guiding modes, and test subjects distribute their preferences for the best guiding mode almost uniformly among the three solutions. From the experiments it also emerges that higher effort is necessary for decoding the sonified instructions if compared to the speech instructions, and that test subjects require frequent `hints' (in the form of speech messages). Despite this, more than 2/3 of test subjects prefer one of the two guiding modes based on sonification. There are two main reasons for this: firstly, with speech messages it is harder to hear the sound of the environment, and secondly sonified messages convey information about the "quantity" of the expected movement.