Open X-Embodiment
1M+ real-robot trajectories from 20+ institutions.
Who Called? V33DA: A Physically Verified Multimodal Benchmark for Vocal Attribution in Zebra Finch Groups Task. Given a detected zebra finch vocalization and the set of birds visible at that moment, determine which bird produced the call. Caller identity is verified physically from on-body accelerometer vibration; the accelerometer channel is withheld from benchmark models and used only by an oracle ceiling. V33DA provides 33,625 verified vocalization events from 10 individually… See the full description on the dataset page:
Source: Hugging Face Hub (songbirdini/v33da). Metadata imported from the dataset’s Hub tags.