Systematic Analysis of Music Representations from BERT
There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fin...
Gespeichert in:
Hauptverfasser: | , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | There have been numerous attempts to represent raw data as numerical vectors
that effectively capture semantic and contextual information. However, in the
field of symbolic music, previous works have attempted to validate their music
embeddings by observing the performance improvement of various fine-tuning
tasks. In this work, we directly analyze embeddings from BERT and BERT with
contrastive learning trained on bar-level MIDI, inspecting their musical
information that can be obtained from MIDI events. We observe that the
embeddings exhibit distinct characteristics of information depending on the
contrastive objectives and the choice of layers. Our code is available at
https://github.com/sjhan91/MusicBERT. |
---|---|
DOI: | 10.48550/arxiv.2306.04628 |