Tempo vs. Pitch: understanding self-supervised tempo estimation

Self-supervision methods learn representations by solving pretext tasks that do not require human-generated labels, alleviating the need for time-consuming annotations. These methods have been applied in computer vision, natural language processing, environmental sound analysis, and recently in musi...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-04
Main Authors: Morais, Giovana, Davies, Matthew E P, Queiroz, Marcelo, Fuentes, Magdalena
Format: Article
Language:English
Subjects:
Annotations
Background noise
Computer vision
Fragility
Information retrieval
Music
Natural language processing
Pitch estimation
Representations
Synthetic data
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Self-supervision methods learn representations by solving pretext tasks that do not require human-generated labels, alleviating the need for time-consuming annotations. These methods have been applied in computer vision, natural language processing, environmental sound analysis, and recently in music information retrieval, e.g. for pitch estimation. Particularly in the context of music, there are few insights about the fragility of these models regarding different distributions of data, and how they could be mitigated. In this paper, we explore these questions by dissecting a self-supervised model for pitch estimation adapted for tempo estimation via rigorous experimentation with synthetic data. Specifically, we study the relationship between the input representation and data distribution for self-supervised tempo estimation.
ISSN:2331-8422
DOI:10.48550/arxiv.2304.06868