vocadito: A dataset of solo vocals with f0, note, and lyric Annotations

vocadito is a dataset of 40 short excerpts of solo, monophonic singing. The excerpts are sung in 7 different languages by singers with varying of levels of training, and are recorded on a variety of devices. For a detailed description, see the technical report.

Annotations are labeled by trained musicians. For each excerpt, we provide:

• frame-level f0 annotations
• 2 versions of note annotations (from 2 different annotators)
• lyrics
• language

Python code for loading this dataset is included in mirdata.