Audio fingerprinting (AFP) allows the identification of unknown audio content by extracting compact representations, termed audio fingerprints, that are designed to remain robust against common audio degradations. Neural AFP methods often employ metric learning, where representation quality is influenced by the nature of the supervision and the utilized loss function. However, recent work unrealistically simulates real-life audio degradation during training, resulting in sub-optimal supervision. Additionally, although several modern metric learning approaches have been proposed, current neural AFP methods continue to rely on the NT‑Xent loss without exploring the recent advances or classical alternatives. In this work, we propose a series of best practices to enhance the self-supervision by leveraging musical signal properties and realistic room acoustics. We then present the first systematic evaluation of various metric learning approaches in the context of AFP, demonstrating that a self‑supervised adaptation of the triplet loss yields superior performance. Our results also reveal that training with multiple positive samples per anchor has critically different effects across loss functions. Our approach is built upon these insights and achieves state-of-the-art performance on both a large, synthetically degraded dataset and a real-world dataset recorded using microphones in diverse music venues.

This work was supported by the pre-doctoral program AGAUR-FI ajuts (2024 FI-3 00065) Joan Oró, funded by the Secretaria d’Universitats i Recerca of the Departament de Recerca i Universitats of the Generalitat de Catalunya; and by the Cátedras ENIA program “IA y Música: Cátedra en Inteligencia Artificial y Música” (TSI-100929-2023-1), funded by the Secretaría de Estado de Digitalización e Inteligencia Artificial and the European Union Next Generation EU. This work was also part of the project TROBA Technologies for the recognition of musical works in the era of dynamic generation of audio content (ACE014/20/000051), within the call Nuclis d’R+D 2024, with the support of ACCIÓ (Agency for Business Competitiveness, Government of Catalonia).

Comunicació presentada al 26th International Society for Music Information Retrieval Conference (ISMIR 2025), celebrada a Daejeon (Korea) del 21 al 25 de setembre del 2025