Overview of the ICASSP 2026 Cadenza Challenge: predicting lyric intelligibility

Roa-Dabike, G., Barker, J.P. orcid.org/0000-0002-1684-5660, Cox, T.J. et al. (8 more authors) (2026) Overview of the ICASSP 2026 Cadenza Challenge: predicting lyric intelligibility. In: ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 03-08 May 2026, Barcelona, Spain. . Institute of Electrical and Electronics Engineers (IEEE), pp. 21757-21759. ISBN: 9798331567026. ISSN: 1520-6149.

Abstract

We present the first open challenge on predicting lyric intelligibility. A new dataset, CLIP1, was introduced, comprising audio samples of popular western music paired with listener intelligibility scores. To model diverse listening profiles, samples were processed with no, mild and moderate simulated hearing loss. A total of 27 systems were submitted by 22 teams. Most systems used foundation models to extract encoder embeddings as high-level acoustic representations, often complemented by signal features and perceptual metrics. Twenty-five systems outperformed the STOI baseline, and 16 outperformed a Whisper-based baseline.

Metadata

Item Type:	Proceedings Paper
Authors/Creators:	Roa-Dabike, G. Barker, J.P. https://orcid.org/0000-0002-1684-5660 Cox, T.J. Akeroyd, M.A. Bannister, S. Fazenda, B. Firth, J. Graetzer, S. Greasley, A. Vos, R.R. Whitmer, W.M.
Copyright, Publisher and Additional Information:	© 2026 The Authors. Except as otherwise noted, this author-accepted version of a journal article published in ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) is made available via the University of Sheffield Research Publications and Copyright Policy under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly cited. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/
Keywords:	hearing loss; machine learning; intelligibility; lyrics; music
Dates:	Published (online): 21 April 2026 Published: 21 April 2026
Institution:	The University of Sheffield
Academic Units:	The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Computer Science (Sheffield)
Funding Information:	Funder Grant number Engineering and Physical Sciences Research Council EP/W019434/1
Date Deposited:	22 May 2026 12:55
Last Modified:	22 May 2026 13:25
Status:	Published
Publisher:	Institute of Electrical and Electronics Engineers (IEEE)
Refereed:	Yes
Identification Number:	10.1109/icassp55912.2026.11463231
Related URLs:	Dataset
Open Archives Initiative ID (OAI ID):	oai:eprints.whiterose.ac.uk:241368

CORE (COnnecting REpositories)

Overview of the ICASSP 2026 Cadenza Challenge: predicting lyric intelligibility

Abstract

Metadata

Download

Accepted Version

Export

Statistics

Overview of the ICASSP 2026 Cadenza Challenge: predicting lyric intelligibility

Abstract

Metadata

Download

Accepted Version

Related datasets

Export

Statistics