Audio-Visual Dataset
H2H Music Improv Dataset
Human-to-Human Free Improvisation with Bidirectional Intention Annotations
Aleksandra Teng Ma, Anthony Cammarota, Jiayi Wang, Alexandria Smith, Cheng-Zhi Anna Huang, Jeffrey Albert, Alexander Lerch
Published in the Proceedings of the International Society for Music Information Retrieval Conference (ISMIR), 2026
Overview
~6 hours of expert duo free improvisations with per-player audio stems and synchronized video. Here we show a selection of clips across musician pairings.
Annotations
Bidirectional: each performer annotated their own intentions and their perception of their partner's actions using our proposed communication model.
How to use
Select Mixture or an individual stem, then hover to preview a moment and click to play from there. Bars show improvisation states from each player's perspective.
Loading annotation data…
Cite
@inproceedings{ma2026h2h,
  title     = {H2H Music Improv: Human-to-Human Free Improvisation with Bidirectional Intention Annotations},
  author    = {Ma, Aleksandra Teng and Cammarota, Anthony and Wang, Jiayi and Smith, Alexandria and Huang, Cheng-Zhi Anna and Albert, Jeffrey and Lerch, Alexander},
  booktitle = {Proceedings of the International Society for Music Information Retrieval Conference (ISMIR)},
  year      = {2026}
}