Listener Acoustic Personalisation (LAP) Challenge
A benchmarking campaign on acoustic and non-acoustic factors in immersive technologies.
Personalized Head-Related Transfer Functions (HRTFs) have shown promise in enhancing auditory localization and immersion in mixed realities. However, relevant issues such as the accurate acquisition of user-specific anatomical data, efficient simulation algorithms, and effective user validation do not converge into a common and internationally recognized benchmark for evaluating HRTFs.
The LAP Challenge endeavours to provide a platform where researchers can explore these challenges, advance the state of the art, and contribute to the development of standardized metrics for personalised spatial audio.
The inaugural edition of the challenge will concentrate on two fundamental aspects of HRTF: spatial sampling and interpolation. Teams are challenged to submit their solutions that address one of two tasks:
- Task 1: HRTF normalisation for merging different HRTF datasets
- Task 2: spatial upsampling for obtaining a high-spatial-resolution HRTF from a very low number of directions
Workshop
The LAP Challenge workshop and award ceremony was hosted by the 32nd European Signal Processing Conference (EUSIPCO 24) on August 29, 2024 – in Lyon, France.
Detailed program: Thursday 29th August, 16:10-17:50 (GMT+2), Room Saint Clair 4
16:10 – 16:20 Welcome and Motivation for the Listener Acoustic Personalization Challenge 2024 Aidan Hogg and Michele Geronazzo
16:20 – 17:40 Main Session
16:20 – 17:00 Task 1: HRTF normalisation for merging different HRTF datasets
16:20 – 16:25: Task 1 Overview and Rules
16:25 – 16:45: Talk from Task 1 Winners: Normalization of Head-Related Transfer Functions Based on Neural Networks. Jiale Zhao, Dingding Yao, Zelin Qiu, Chengzhong Wang, and Junfeng Li
16:45 – 17:00: Task 1 Full Results and Discussion
17:00 – 17:40 Task 2: Spatial upsampling for obtaining a high-spatial-resolution HRTF from a very low number of directions
17:00 – 17:05: Task 2 Overview and Rules
17:05 – 17:25: Talk from Task 2 Winners: Retrieval-Augmented Neural Field for HRTF Upsampling and Personalization. Yoshiki Masuyama, Gordon Wichern, Francois G. Germain, Christopher Ick, and Jonathan Le Roux
17:25 – 17:40: Task 2 Full Results and Discussion
17:40 – 17:50 Award Ceremony and Closing
Task description documents
The documents below outline the complete description for each task.
The evaluation code is also publicly available here:
Results
A draft technical report covering the full results of the challenge is available here (a complete report and DOI will be available soon):
Submissions
To submit your solutions for the LAP Challenge tasks, fill out the forms below:
Key Dates
Task specifics released: March 27, 2024Evaluation code released: April 4, 2024Submission deadline: July 1, 2024Award ceremony: August 29, 2024
Editorial aspects
Top-ranked solutions will be invited to submit a paper describing their method and result to be published in the IEEE Open Journal of Signal Processing (OJ-SP), a fully open-access publication of the IEEE Signal Processing Society, with a scope encompassing the full range of technical activities of the Society. Article processing charge (APC) waivers will be also available upon request.
Since the challenge provides original data and evaluation processes, participants also have the flexibility to submit already published solutions.
Awards
Winners of the two tasks will be invited to participate and reimbursed for the registration fee to EUSIPCO24 and, until available funds are reached, for travel and accommodation.
Organisers
Chair: Michele Geronazzo, University of Padova, IT, and Imperial College London, UK
Co-chair: Lorenzo Picinali, Imperial College London, UK
Chair of the Implementation: Roberto Barumerli, University of Verona, IT,
Chair of the Web Site and Dissemination: Aidan Hogg, Queen Mary University of London, UK
Fabian Brinkmann, Technische Universität Berlin, DE
Glen McLachlan, University of Antwerp, BE
Stavros Ntalampiras, University of Milan, IT
Johan Pauwels, Queen Mary University of London, UK
Katarina Poole, Imperial College London, UK
Rapolas Daugintis, Imperial College London, UK
Our Sponsors
