Two-level Explanations in Music Emotion Recognition

Verena Haunschmid; Shreyan Chowdhury; Gerhard Widmer

Two-level Explanations in Music Emotion Recognition

Verena Haunschmid, Shreyan Chowdhury, Gerhard Widmer

Institute of Computational Perception

Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review

Abstract

Current ML models for music emotion recognition, while generally working quite well, do not give meaningful or intuitive explanations for their predictions. In this work, we propose a 2-step procedure to arrive at spectrogram-level explanations that connect certain aspects of the audio to interpretable mid-level perceptual features, and these to the actual emotion prediction. That makes it possible to focus on specific musical reasons for a prediction (in terms of perceptual features), and to trace these back to patterns in the audio that can be interpreted visually and acoustically.

Original language	English
Title of host publication	Proceedings of the 36 th International Conference on Machine Learning
Number of pages	3
Publication status	Published - Jun 2019

Fields of science

202002 Audiovisual media
102 Computer Sciences
102001 Artificial intelligence
102003 Image processing
102015 Information systems

JKU Focus areas

Digital Transformation

Con Espressione - Getting at the Heart of Things: Towards Expressivity-aware Computer Systems in Music (ERC Advanced Grant)
Widmer, G. (PI)
01.01.2016 → 31.12.2021
Project: Funded research › EU - European Union

Cite this

@inproceedings{91818d7783974f92a2a29f39e910b99c,

title = "Two-level Explanations in Music Emotion Recognition",

abstract = "Current ML models for music emotion recognition, while generally working quite well, do not give meaningful or intuitive explanations for their predictions. In this work, we propose a 2-step procedure to arrive at spectrogram-level explanations that connect certain aspects of the audio to interpretable mid-level perceptual features, and these to the actual emotion prediction. That makes it possible to focus on specific musical reasons for a prediction (in terms of perceptual features), and to trace these back to patterns in the audio that can be interpreted visually and acoustically.",

author = "Verena Haunschmid and Shreyan Chowdhury and Gerhard Widmer",

year = "2019",

month = jun,

language = "English",

booktitle = "Proceedings of the 36 th International Conference on Machine Learning",

}

TY - GEN

T1 - Two-level Explanations in Music Emotion Recognition

AU - Haunschmid, Verena

AU - Chowdhury, Shreyan

AU - Widmer, Gerhard

PY - 2019/6

Y1 - 2019/6

N2 - Current ML models for music emotion recognition, while generally working quite well, do not give meaningful or intuitive explanations for their predictions. In this work, we propose a 2-step procedure to arrive at spectrogram-level explanations that connect certain aspects of the audio to interpretable mid-level perceptual features, and these to the actual emotion prediction. That makes it possible to focus on specific musical reasons for a prediction (in terms of perceptual features), and to trace these back to patterns in the audio that can be interpreted visually and acoustically.

AB - Current ML models for music emotion recognition, while generally working quite well, do not give meaningful or intuitive explanations for their predictions. In this work, we propose a 2-step procedure to arrive at spectrogram-level explanations that connect certain aspects of the audio to interpretable mid-level perceptual features, and these to the actual emotion prediction. That makes it possible to focus on specific musical reasons for a prediction (in terms of perceptual features), and to trace these back to patterns in the audio that can be interpreted visually and acoustically.

M3 - Conference proceedings

BT - Proceedings of the 36 th International Conference on Machine Learning

ER -

Two-level Explanations in Music Emotion Recognition

Abstract

Fields of science

JKU Focus areas

Projects

Con Espressione - Getting at the Heart of Things: Towards Expressivity-aware Computer Systems in Music (ERC Advanced Grant)

Cite this