Zum Inhalt springen

Nur-Lesen-Fassung: Das Forschungsinformationssystem (FIS) ist aus Sicherheitsgründen vorübergehend durch diese Nur-Lesen-Fassung ersetzt. Alle öffentlichen Inhalte und Volltexte sind weiterhin abrufbar; Anmeldung, Bearbeitung, Einreichung und Statistiken stehen nicht zur Verfügung. Bei Fragen: fis@uni-bamberg.de

Start › Publikationen › How Robust are Audio Embeddings for Polyphonic Sound Event Tagging?

Publikation

How Robust are Audio Embeddings for Polyphonic Sound Event Tagging?

Abeßer, Jakob; Grollmisch, Sascha; Müller, Meinard (2026): How Robust are Audio Embeddings for Polyphonic Sound Event Tagging?, in: Bamberg: Otto-Friedrich-Universität, S. 2658–2667.

Details

Fakultät/Lehrstuhl
Autorin(nen)/Autor(en)
Grollmisch, Sascha
Müller, Meinard
Verlagsangabe
Otto-Friedrich-Universität : Bamberg
Erscheinungsjahr
2026
Quelle/Weitere Ausgabe
IEEE ACM transactions on audio, speech, and language processing : TASLP, New York, NY: IEEE, 2023, Jg. 31, S. 2658–2667, ISSN: 2329-9290
Jahr der Erstpublikation
2023
Sprache(n)
Englisch
Lizenz
URN
Abstract

Sound classification algorithms are challenged by the natural variability of everyday sounds, particularly for large sound class taxonomies. In order to be applicable in real-life environments, such algorithms must also be able to handle polyphonic scenarios, where simultaneously occurring and overlapping sound events need to be classified. With the rapid progress of deep learning, several deep audio embeddings (DAEs) have been proposed as pre-trained feature representations for sound classification. In this article, we analyze the embedding spaces of two non-trainable audio representations (NTARs) and five DAEs for sound classification in polyphonic scenarios (sound event tagging) and make several contributions. First, we compare general properties like the inter-correlation between feature dimensions and the scattering of sound classes in the embedding spaces. Second, we test the robustness of the embeddings against several audio degradations and propose two sensitivity measures based on a class-agnostic and a class-centric view on the resulting drift in the embedding space. Finally, as a central contribution, we study how a blending between pairs of sounds maps to embedding space trajectories and how the path of these trajectories can cause classification errors due to their proximity to other sound classes. Throughout our analyses, the PANN embeddings have shown the best overall performance for low-polyphony sound event tagging.

Freie Schlagwort(e)
sound event tagging
sound polyphony
deep audio embeddings
embedding space
Peer Reviewed
ja
Internationale Verbreitung
ja
Open-Access-Zeitschrift
nein
Dokumenttyp
Artikel in einer Zeitschrift / Postprint
URI
Datum der Freischaltung
02.02.2026

Permalink: https://fis.uni-bamberg.de/handle/uniba/112893 · Daten als JSON · Zitieren: BibTeX · RIS

Datenstand: 16.11.2026

Eingeschränkter Lesezugriff: Anmeldung, Bearbeitung und Einreichung sind vorübergehend deaktiviert.