Skip to content

Read-only version: For security reasons, the Research Information System (FIS) has been temporarily replaced by this read-only version. All public content and full texts remain available; login, editing, submission and statistics are not available. Questions: fis@uni-bamberg.de

Home › Publications › How Reliable are Posterior Class Probabilties in Automatic Music Classification?

Publication

How Reliable are Posterior Class Probabilties in Automatic Music Classification?

Lukashevich, Hanna; Grollmisch, Sascha; Abeßer, Jakob; u. a. (2026): How Reliable are Posterior Class Probabilties in Automatic Music Classification?, in: Bamberg: Otto-Friedrich-Universität, S. 45–50.

Details

Faculty/Chair
Faculty/Professorship
Author
Lukashevich, Hanna
Grollmisch, Sascha
Stober, Sebastian
Bös, Joachim
Publisher Information
Otto-Friedrich-Universität : Bamberg
Year of publication
2026
Source/Other editions
AM ’23: Proceedings of the 18th International Audio Mostly Conference, ACM, 2023, S. 45–50, ISBN: 979-8-4007-0818-3
Year of first publication
2023
Language
English
Licence
URN
Abstract

Music classification algorithms use signal processing and machine learning approaches to extract and enrich metadata for audio recordings in music archives. Common tasks include music genre classification, where each song is assigned a single label (such as Rock, Pop, or Jazz), and musical instrument classification. Since music metadata can be ambiguous, classification algorithms cannot always achieve fully accurate predictions. Therefore, our focus extends beyond the correctly estimated class labels to include realistic confidence values for each potential genre or instrument label. In practice, many state-of-the-art classification algorithms based on deep neural networks exhibit overconfident predictions, complicating the interpretation of the final output values. In this work, we examine whether the issue of overconfident predictions and, consequently, non-representative confidence values is also relevant to music genre classification and musical instrument classification. Moreover, we describe techniques to mitigate this behavior and assess the impact of deep ensembles and temperature scaling in generating more realistic confidence outputs, which can be directly employed in real-world music tagging applications.

Keywords
music information retrieval
music classifcation
uncertainty
temperature scaling
deep ensembles
Peer Reviewed
yes
International Distribution
yes
Type
Conferenceobject
URI
Activation date
2026-02-02

Permalink: https://fis.uni-bamberg.de/handle/uniba/112894 · Data as JSON · Cite: BibTeX · RIS

Data as of: 2026-11-16

Restricted read-only access: login, editing and submission are temporarily disabled.