Text augmentations from music artist biographies improve performance on genre classification, emotion recognition and music tagging

Rossetto, Federico (2026) Text augmentations from music artist biographies improve performance on genre classification, emotion recognition and music tagging. PhD thesis, University of Glasgow.

Abstract

Music Information Retrieval (MIR) aims to develop automated systems to analyse, categorise, and recommend music. Multimodal learning plays an important but challenging role in MIR. Compared to more common multimodal settings such as speech transcription or image captioning, aligning audio with textual descriptions of music is particularly difficult because many musical concepts are abstract, culturally grounded, and not directly observable in the signal.
Several approaches have explored incorporating textual information into MIR systems, leveraging lyrics, editorial metadata, or specialised audio-text corpora. However, these approaches exhibit a common trade-off between the semantic richness of the textual source and its availability and reproducibility, often due to coverage limitations or copyright restrictions.
In this thesis, we investigate artist biographies as an alternative source of textual supervision. We show that artist biographies provide a scalable and reproducible source of high-level semantic information. To this end, we construct two corpora: one derived from web-scraped, community-authored content, and one generated using a large language model. Both are made available for reuse, with a documented reconstruction procedure for the web-scraped text, whose licensing prevents direct redistribution. We further introduce a unified evaluation framework that enables controlled comparisons of representations across multiple semantic MIR tasks and allows us to assess the contributions of textual, audio, and multimodal inputs, both in isolation and in combination.
Experimental results show that incorporating artist biographies improves performance across most evaluated tasks. Relative to audio-only baselines, multimodal representations yield improvements of up to +38% in genre classification, +28% in emotion recognition, and +14% in music tagging, although one smaller dataset remains inconclusive due to limited training data. Additionally, we provide initial evidence that models can internalise the semantic information in biographies through generative pretraining, leading to improved performance even when textual inputs are unavailable at inference time.
These findings establish artist biographies as an effective, scalable, and reproducible source of semantic supervision for multimodal MIR and highlight the broader potential of structured textual knowledge for enhancing music representation learning.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: This work was partially funded by MoodAgent.
• Subjects: M Music and Books on Music > M Music; Q Science > QA Mathematics > QA75 Electronic computers. Computer science; Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources
• Colleges/Schools: College of Science and Engineering > School of Computing Science
• Funder's Name: MoodAgent
• Supervisor's Name: Murray-Smith, Professor Roderick
• Date of Award: 2026
• Depositing User: Theses Team
• Unique ID: glathesis:2026-86297
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 05 Nov 2026 14:03
• Last Modified: 05 Nov 2026 14:04
• Thesis DOI: 10.5525/gla.thesis.86297
• URI: https://theses.gla.ac.uk/id/eprint/86297

Related URLs

• Conference proceeding