Saved in:
Bibliographic Details
Main Authors: Shin, Saebyeol, Wan, Chao, Liu, Zhenzhen, Lovelace, Justin, Lin, Daniel C., Weinberger, Kilian Q., Thickstun, John
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.24193
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911710831443968
author Shin, Saebyeol
Wan, Chao
Liu, Zhenzhen
Lovelace, Justin
Lin, Daniel C.
Weinberger, Kilian Q.
Thickstun, John
author_facet Shin, Saebyeol
Wan, Chao
Liu, Zhenzhen
Lovelace, Justin
Lin, Daniel C.
Weinberger, Kilian Q.
Thickstun, John
contents Competitive music transcription models require large amounts of paired audio-score data, which is scarce due to collection costs, alignment difficulty, and copyright restrictions. Meanwhile, vast quantities of unpaired audio recordings and symbolic scores are freely available but have gone unused. We adopt a cycle-consistent translation framework in which a small amount of paired data acts as a minimal anchor, unlocking the full potential of the unpaired pool. We find that: unpaired data yields surprisingly large gains, especially under limited supervision; unpaired audio contributes more than unpaired scores; incorporating unlabeled audio from a new instrument during training improves transcription for that instrument without any paired supervision. Together, these results suggest that scaling unpaired data offers a practical path toward high-quality transcription for instruments where labeled data remains scarce.
format Preprint
id arxiv_https___arxiv_org_abs_2605_24193
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Music Transcription with (Almost) No Supervision
Shin, Saebyeol
Wan, Chao
Liu, Zhenzhen
Lovelace, Justin
Lin, Daniel C.
Weinberger, Kilian Q.
Thickstun, John
Sound
Machine Learning
Competitive music transcription models require large amounts of paired audio-score data, which is scarce due to collection costs, alignment difficulty, and copyright restrictions. Meanwhile, vast quantities of unpaired audio recordings and symbolic scores are freely available but have gone unused. We adopt a cycle-consistent translation framework in which a small amount of paired data acts as a minimal anchor, unlocking the full potential of the unpaired pool. We find that: unpaired data yields surprisingly large gains, especially under limited supervision; unpaired audio contributes more than unpaired scores; incorporating unlabeled audio from a new instrument during training improves transcription for that instrument without any paired supervision. Together, these results suggest that scaling unpaired data offers a practical path toward high-quality transcription for instruments where labeled data remains scarce.
title Music Transcription with (Almost) No Supervision
topic Sound
Machine Learning
url https://arxiv.org/abs/2605.24193